list_key,line_no,section_path,title,url,arxiv_id,venue,year,month,date_basis,source_type,description meirtz_context,165,Related Survey,A Survey of Large Language Models,https://arxiv.org/abs/2303.18223,2303.18223,arXiv,2023,3,arxiv_id,academic, meirtz_context,170,Related Survey,The Prompt Report: A Systematic Survey of Prompt Engineering Techniques,https://arxiv.org/abs/2406.06608,2406.06608,arXiv,2024,6,arxiv_id,academic, meirtz_context,174,Related Survey,A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applications,https://arxiv.org/abs/2402.07927,2402.07927,arXiv,2024,2,arxiv_id,academic, meirtz_context,176,Related Survey,A Systematic Survey of Prompt Engineering on Vision-Language Foundation Models,https://arxiv.org/abs/2307.12980,2307.12980,arXiv,2023,7,arxiv_id,academic, meirtz_context,185,Related Survey,A Survey on In-context Learning,https://doi.org/10.18653/v1/2024.emnlp-main.64,,EMNLP,2024,11,badge,academic, meirtz_context,189,Related Survey,The Mystery of In-Context Learning: A Comprehensive Survey on Interpretation and Analysis,https://arxiv.org/abs/2311.00237,2311.00237,arXiv,2023,11,arxiv_id,academic, meirtz_context,193,Related Survey,"A Comprehensive Survey of Retrieval-Augmented Generation (RAG): Evolution, Current Landscape and Future Directions",https://arxiv.org/abs/2410.12837,2410.12837,arXiv,2024,10,arxiv_id,academic, meirtz_context,195,Related Survey,Retrieval-Augmented Generation for Large Language Models: A Survey,https://arxiv.org/abs/2312.10997,2312.10997,arXiv,2023,12,arxiv_id,academic, meirtz_context,199,Related Survey,A Survey on Knowledge-Oriented Retrieval-Augmented Generation,https://arxiv.org/abs/2503.10677,2503.10677,arXiv,2025,3,arxiv_id,academic, meirtz_context,209,Related Survey,A Survey on the Memory Mechanism of Large Language Model based Agents,https://arxiv.org/abs/2404.13501,2404.13501,arXiv,2024,4,arxiv_id,academic, meirtz_context,213,Related Survey,Survey on Memory-Augmented Neural Networks: Cognitive Insights to AI Applications,https://arxiv.org/abs/2312.06141,2312.06141,arXiv,2023,12,arxiv_id,academic, meirtz_context,215,Related Survey,From Human Memory to AI Memory: A Survey on Memory Mechanisms in the Era of LLMs,https://arxiv.org/abs/2504.15965,2504.15965,arXiv,2025,4,arxiv_id,academic, meirtz_context,217,Related Survey,Survey on Evaluation of LLM-based Agents,https://arxiv.org/abs/2503.16416,2503.16416,arXiv,2025,3,arxiv_id,academic, meirtz_context,219,Related Survey,A Survey of Personalized Large Language Models: Progress and Future Directions,https://arxiv.org/abs/2502.11528,2502.11528,arXiv,2025,2,arxiv_id,academic, meirtz_context,221,Related Survey,Agentic Retrieval-Augmented Generation: A Survey,https://arxiv.org/abs/2501.09136,2501.09136,arXiv,2025,1,arxiv_id,academic, meirtz_context,223,Related Survey,Retrieval-Augmented Generation with Graphs (GraphRAG),https://arxiv.org/abs/2501.00309,2501.00309,arXiv,2025,1,arxiv_id,academic, meirtz_context,227,Related Survey,The Landscape of Agentic Reinforcement Learning for LLMs: A Survey,https://arxiv.org/abs/2509.02547,2509.02547,arXiv,2025,9,arxiv_id,academic, meirtz_context,235,Related Survey,Evaluating Very Long-Term Conversational Memory of LLM Agents (LOCOMO),https://arxiv.org/abs/2402.17753,2402.17753,ACL,2024,2,arxiv_id,academic, meirtz_context,239,Related Survey,Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactions,https://arxiv.org/abs/2507.05257,2507.05257,arXiv,2025,7,arxiv_id,academic, meirtz_context,245,Related Survey,Episodic Memories Generation and Evaluation Benchmark for Large Language Models,https://arxiv.org/abs/2501.13121,2501.13121,arXiv,2025,1,arxiv_id,academic, meirtz_context,247,Related Survey,On the Structural Memory of LLM Agents,https://arxiv.org/abs/2412.15266,2412.15266,arXiv,2024,12,arxiv_id,academic, meirtz_context,249,Related Survey,"HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering",https://arxiv.org/abs/1809.09600,1809.09600,EMNLP,2018,9,arxiv_id,academic, meirtz_context,257,Related Survey,Neural Turing Machines,https://arxiv.org/abs/1410.5401,1410.5401,arXiv,2014,10,arxiv_id,academic, meirtz_context,259,Related Survey,Differentiable Neural Computers,https://arxiv.org/abs/1610.06258,1610.06258,arXiv,2016,10,arxiv_id,academic, meirtz_context,263,Related Survey,A Brain-inspired Memory Transformation based Differentiable Neural Computer,https://arxiv.org/abs/2301.02809,2301.02809,arXiv,2023,1,arxiv_id,academic, meirtz_context,265,Related Survey,Differentiable Neural Computers with Memory Demon,https://arxiv.org/abs/2211.02987,2211.02987,arXiv,2022,11,arxiv_id,academic, meirtz_context,271,Related Survey,Memorizing Transformers,https://arxiv.org/abs/2203.08913,2203.08913,arXiv,2022,3,arxiv_id,academic, meirtz_context,273,Related Survey,Recurrent Memory Transformer,https://arxiv.org/abs/2207.06881,2207.06881,NeurIPS,2022,7,arxiv_id,academic, meirtz_context,277,Related Survey,Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention,https://arxiv.org/abs/2404.07143,2404.07143,arXiv,2024,4,arxiv_id,academic, meirtz_context,279,Related Survey,Memformer: A Memory-Augmented Transformer for Sequence Modeling,https://arxiv.org/abs/2010.06891,2010.06891,arXiv,2020,10,arxiv_id,academic, meirtz_context,281,Related Survey,Token Turing Machines,https://arxiv.org/abs/2211.09119,2211.09119,arXiv,2022,11,arxiv_id,academic, meirtz_context,283,Related Survey,TransformerFAM: Feedback Attention is Working Memory,https://arxiv.org/abs/2404.09173,2404.09173,arXiv,2024,4,arxiv_id,academic, meirtz_context,289,Related Survey,MemGPT: Towards LLMs as Operating Systems,https://arxiv.org/abs/2310.08560,2310.08560,arXiv,2023,10,arxiv_id,academic, meirtz_context,293,Related Survey,MemoryBank: Enhancing Large Language Models with Long-Term Memory,https://arxiv.org/abs/2305.10250,2305.10250,arXiv,2023,5,arxiv_id,academic, meirtz_context,297,Related Survey,MEM0: Building Production-Ready AI Agents with Scalable Long-Term Memory,https://arxiv.org/abs/2504.19413,2504.19413,arXiv,2025,4,arxiv_id,academic, meirtz_context,301,Related Survey,MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agents,https://arxiv.org/abs/2506.15841,2506.15841,arXiv,2025,6,arxiv_id,academic, meirtz_context,305,Related Survey,A-MEM: Agentic Memory for LLM Agents,https://arxiv.org/abs/2502.12110,2502.12110,arXiv,2025,2,arxiv_id,academic, meirtz_context,309,Related Survey,MemAgent: Reshaping Long-Context LLM with Multi-Conv RL-based Memory Agent,https://arxiv.org/abs/2507.02259,2507.02259,arXiv,2025,7,arxiv_id,academic, meirtz_context,311,Related Survey,Memory OS of AI Agent,https://arxiv.org/abs/2506.06326,2506.06326,arXiv,2025,6,arxiv_id,academic, meirtz_context,319,Related Survey,arigraph: learning knowledge graph world models with episodic memory for llm agents,https://arxiv.org/abs/2407.04363,2407.04363,arXiv,2024,7,arxiv_id,academic, meirtz_context,321,Related Survey,Zep: A Temporal Knowledge Graph Architecture for Agent Memory,https://arxiv.org/abs/2501.13956,2501.13956,arXiv,2025,1,arxiv_id,academic, meirtz_context,325,Related Survey,KG-Agent: An Efficient Autonomous Agent Framework for Complex Reasoning over Knowledge Graph,https://arxiv.org/abs/2402.11163,2402.11163,arXiv,2024,2,arxiv_id,academic, meirtz_context,327,Related Survey,GraphReader: Building Graph-based Agent to Enhance Long-Context Abilities of Large Language Models,https://arxiv.org/abs/2406.14550,2406.14550,arXiv,2024,6,arxiv_id,academic, meirtz_context,329,Related Survey,From Local to Global: A GraphRAG Approach to Query-Focused Summarization,https://arxiv.org/abs/2404.16130,2404.16130,arXiv,2024,4,arxiv_id,academic, meirtz_context,333,Related Survey,Knowledge Graph-Guided Retrieval Augmented Generation,https://arxiv.org/abs/2502.06864,2502.06864,arXiv,2025,2,arxiv_id,academic, meirtz_context,339,Related Survey,Larimar: Large Language Models with Episodic Memory Control,https://arxiv.org/abs/2403.11901,2403.11901,ICML,2024,3,arxiv_id,academic, meirtz_context,341,Related Survey,EM-LLM: Human-like Episodic Memory for Infinite Context LLMs,https://arxiv.org/abs/2407.09450,2407.09450,ICLR,2024,7,arxiv_id,academic, meirtz_context,345,Related Survey,Large Language Models with Controllable Working Memory,https://arxiv.org/abs/2211.05110,2211.05110,arXiv,2022,11,arxiv_id,academic, meirtz_context,347,Related Survey,Empowering Working Memory for Large Language Model Agents,https://arxiv.org/abs/2312.17259,2312.17259,arXiv,2023,12,arxiv_id,academic, meirtz_context,353,Related Survey,MemoChat: Tuning LLMs to Use Memos for Consistent Long-Range Open-Domain Conversation,https://arxiv.org/abs/2308.08239,2308.08239,arXiv,2023,8,arxiv_id,academic, meirtz_context,355,Related Survey,Think-in-Memory: Recalling and Post-thinking Enable LLMs with Long-Term Memory,https://arxiv.org/abs/2311.08719,2311.08719,arXiv,2023,11,arxiv_id,academic, meirtz_context,357,Related Survey,Generative Agents: Interactive Simulacra of Human Behavior,https://arxiv.org/abs/2304.03442,2304.03442,arXiv,2023,4,arxiv_id,academic, meirtz_context,359,Related Survey,Self-Controlled Memory Framework for Large Language Models,https://arxiv.org/abs/2304.13343,2304.13343,arXiv,2023,4,arxiv_id,academic, meirtz_context,390,Related Survey,Retrieval-Augmented Generation for AI-Generated Content: A Survey,https://arxiv.org/abs/2402.19473,2402.19473,arXiv,2024,2,arxiv_id,academic, meirtz_context,394,Related Survey,Retrieval Augmented Generation (RAG) and Beyond: A Comprehensive Survey on How to Make your LLMs use External Data More Wisely,https://arxiv.org/abs/2409.14924,2409.14924,arXiv,2024,9,arxiv_id,academic, meirtz_context,510,Related Blogs,Bernstein,https://github.com/sipyourdrink-ltd/bernstein,,,,,,repository,"Multi-agent orchestrator with context-aware model routing. Routes tasks to optimal models based on complexity, manages context across parallel agents." meirtz_context,511,Related Blogs,"The rise of ""context engineering""",https://blog.langchain.com/the-rise-of-context-engineering/,,,,,,grey, meirtz_context,512,Related Blogs,"The New Skill in AI is Not Prompting, It's Context Engineering",https://www.philschmid.de/context-engineering,,,,,,grey, meirtz_context,513,Related Blogs,"davidkimai/Context-Engineering: ""Context engineering is the delicate art and science of filling the context window with just the right information for the next step.""",https://github.com/davidkimai/Context-Engineering,,,,,,repository, meirtz_context,514,Related Blogs,"Context Engineering is Runtime of AI Agents | by Bijit Ghosh | Jun, 2025 | Medium",https://medium.com/@bijit211987/context-engineering-is-runtime-of-ai-agents-411c9b2ef1cb,,,,,,grey, meirtz_context,515,Related Blogs,Context Engineering,https://blog.langchain.com/context-engineering-for-agents/,,,,,,grey, meirtz_context,516,Related Blogs,Context Engineering for Agents,https://rlancemartin.github.io/2025/06/23/context_engineering/,,,2025,6,url_date,grey, meirtz_context,517,Related Blogs,Cognition | Don't Build Multi-Agents,https://cognition.ai/blog/dont-build-multi-agents,,,,,,grey, meirtz_context,518,Related Blogs,从Prompt Engineering到Context Engineering - 53AI-AI知识库|大模型知识库|大模型训练|智能体开发,https://www.53ai.com/news/tishicikuangjia/2025062727685.html,,,,,,grey, meirtz_context,519,Related Blogs,What Is Context Engineering?,https://www.outcomeops.ai/context-engineering,,,,,,grey, meirtz_context,523,Related Blogs > Social Media & Talks,Mastering Claude Code in 30 minutes,https://www.youtube.com/watch?v=6eBSHbLKuN0,,,,,,grey, meirtz_context,524,Related Blogs > Social Media & Talks,Context Engineering for Agents,https://www.youtube.com/watch?v=4GiqzUHD5AA,,,,,,grey, meirtz_context,525,Related Blogs > Social Media & Talks,"Andrej Karpathy on X: ""+1 for ""context engineering"" over ""prompt engineering""",https://x.com/karpathy/status/1937902205765607626?ref=blog.langchain.com,,,,,,grey, meirtz_context,526,Related Blogs > Social Media & Talks,复旦大学/上海创智学院邱锡鹏:Context Scaling,通往AGI的下一幕,https://mp.weixin.qq.com/s/Knej0qbyr5j5KX_BO7FGew,,,,,,grey, meirtz_context,629,"Components, Techniques and Architectures > Context Scaling",Extending Context Window of Large Language Models via Position Interpolation,https://arxiv.org/abs/2306.15595,2306.15595,arXiv,2023,6,arxiv_id,academic, meirtz_context,633,"Components, Techniques and Architectures > Context Scaling",YaRN: Efficient Context Window Extension of Large Language Models,https://arxiv.org/abs/2309.00071,2309.00071,ICLR,2023,9,arxiv_id,academic, meirtz_context,637,"Components, Techniques and Architectures > Context Scaling",LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens,https://arxiv.org/abs/2402.13753,2402.13753,ICML,2024,2,arxiv_id,academic, meirtz_context,649,"Components, Techniques and Architectures > Context Scaling",Fast Multipole Attention: A Divide-and-Conquer Attention Mechanism for Long Sequences,https://arxiv.org/abs/2310.11960,2310.11960,ICLR,2023,10,arxiv_id,academic, meirtz_context,653,"Components, Techniques and Architectures > Context Scaling",Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention,https://arxiv.org/abs/2404.07143,2404.07143,arXiv,2024,4,arxiv_id,academic, meirtz_context,661,"Components, Techniques and Architectures > Context Scaling",Star Attention: Efficient LLM Inference over Long Sequences,https://arxiv.org/abs/2411.17116,2411.17116,arXiv,2024,11,arxiv_id,academic, meirtz_context,669,"Components, Techniques and Architectures > Context Scaling",TokenSwift: Lossless Acceleration of Ultra Long Sequence Generation,https://arxiv.org/abs/2502.18890,2502.18890,ICML,2025,2,arxiv_id,academic, meirtz_context,677,"Components, Techniques and Architectures > Context Scaling",∞Bench: Extending Long Context Evaluation Beyond 100K Tokens,https://arxiv.org/abs/2412.00359,2412.00359,ACL,2024,12,arxiv_id,academic, meirtz_context,685,"Components, Techniques and Architectures > Context Scaling",Beyond the Limits: A Survey of Techniques to Extend the Context Length in Large Language Models,https://arxiv.org/abs/2402.02244,2402.02244,arXiv,2024,2,arxiv_id,academic, meirtz_context,687,"Components, Techniques and Architectures > Context Scaling",A Controlled Study on Long Context Extension and Generalization in LLMs,https://arxiv.org/abs/2409.12181,2409.12181,arXiv,2024,9,arxiv_id,academic, meirtz_context,699,"Components, Techniques and Architectures > Context Scaling",Towards LLM-Centric Multimodal Fusion: A Survey on Integration Strategies and Techniques,https://arxiv.org/abs/2506.04788,2506.04788,arXiv,2025,6,arxiv_id,academic, meirtz_context,701,"Components, Techniques and Architectures > Context Scaling",Browse and Concentrate: Comprehending Multimodal Content via Prior-LLM Context Fusion,https://doi.org/10.18653/v1/2024.acl-long.605,,ACL,2024,8,badge,academic, meirtz_context,705,"Components, Techniques and Architectures > Context Scaling",V2PE: Improving Multimodal Long-Context Capability of Vision-Language Models with Variable Visual Position Encoding,https://arxiv.org/abs/2412.09616,2412.09616,arXiv,2024,12,arxiv_id,academic, meirtz_context,709,"Components, Techniques and Architectures > Context Scaling",Flamingo: a Visual Language Model for Few-Shot Learning,https://arxiv.org/abs/2204.14198,2204.14198,NeurIPS,2022,4,arxiv_id,academic, meirtz_context,713,"Components, Techniques and Architectures > Context Scaling",LensVLM: Selective Context Expansion for Compressed Visual Representation of Text,https://arxiv.org/abs/2605.07019,2605.07019,arXiv,2026,5,arxiv_id,academic, meirtz_context,724,"Components, Techniques and Architectures > Context Scaling",AVicuna: Audio-Visual LLM with Interleaver and Context-Boundary Alignment for Temporal Referential Dialogue,https://arxiv.org/abs/2403.16276,2403.16276,arXiv,2024,3,arxiv_id,academic, meirtz_context,726,"Components, Techniques and Architectures > Context Scaling",SonicVisionLM: Playing Sound with Vision Language Models,https://arxiv.org/abs/2401.04394,2401.04394,CVPR,2024,1,arxiv_id,academic, meirtz_context,730,"Components, Techniques and Architectures > Context Scaling",SAVEn-Vid: Synergistic Audio-Visual Integration for Enhanced Understanding in Long Video Context,https://arxiv.org/abs/2411.16213,2411.16213,arXiv,2024,11,arxiv_id,academic, meirtz_context,742,"Components, Techniques and Architectures > Context Scaling",CaMML: Context-Aware Multimodal Learner for Large Models,https://arxiv.org/abs/2404.11406,2404.11406,ACL,2024,4,arxiv_id,academic, meirtz_context,746,"Components, Techniques and Architectures > Context Scaling",CAMA: Enhancing Multimodal In-Context Learning with Context-Aware Modulated Attention,https://arxiv.org/abs/2505.17097,2505.17097,arXiv,2025,5,arxiv_id,academic, meirtz_context,776,"Components, Techniques and Architectures > Context Scaling",Video Understanding with Large Language Models: A Survey,https://arxiv.org/abs/2312.17432,2312.17432,arXiv,2023,12,arxiv_id,academic, meirtz_context,790,"Components, Techniques and Architectures > Context Management in Production",OpenAI Agents Guide,https://platform.openai.com/docs/guides/agents,,OpenAI,,,,grey, meirtz_context,791,"Components, Techniques and Architectures > Context Management in Production","OpenAI Tools: Conversation State, Prompt Caching, and Compaction",https://developers.openai.com/api/docs/guides/tools,,OpenAI,,,,grey, meirtz_context,792,"Components, Techniques and Architectures > Context Management in Production",Google ADK: Context Caching and Context Compression,https://google.github.io/adk-docs/,,Google,,,,grey, meirtz_context,793,"Components, Techniques and Architectures > Context Management in Production",Claude Code Memory and Scoped Project Instructions,https://docs.anthropic.com/en/docs/claude-code/memory,,Anthropic,,,,grey, meirtz_context,794,"Components, Techniques and Architectures > Context Management in Production",LangChain Deep Agents: Filesystem-Based Context Management,https://docs.langchain.com/oss/python/deepagents/overview,,LangChain,,,,grey, meirtz_context,799,"Components, Techniques and Architectures > Context Management in Production",Entroly: The Context Engineering Engine for AI Coding Agents,https://github.com/juyterman1000/entroly,,Tool,,,,repository, meirtz_context,819,"Components, Techniques and Architectures > Structured Data Integration",Learn Together: Joint Multitask Finetuning of Pretrained KG-enhanced LLM for Downstream Tasks,https://doi.org/10.18653/v1/2025.genaik-1.2,,ICCL,2025,1,badge,academic, meirtz_context,825,"Components, Techniques and Architectures > Structured Data Integration",Knowledge Graph-Guided Retrieval Augmented Generation,https://arxiv.org/abs/2502.06864,2502.06864,arXiv,2025,2,arxiv_id,academic, meirtz_context,829,"Components, Techniques and Architectures > Structured Data Integration",KGLA: Knowledge Graph Enhanced Language Agents for Customer Service,https://arxiv.org/abs/2410.19627,2410.19627,arXiv,2024,10,arxiv_id,academic, meirtz_context,836,"Components, Techniques and Architectures > Structured Data Integration",Are Large Language Models In-Context Graph Learners?,https://arxiv.org/abs/2502.13562,2502.13562,arXiv,2025,2,arxiv_id,academic, meirtz_context,840,"Components, Techniques and Architectures > Structured Data Integration",Let's Ask GNN: Empowering Large Language Model for Graph In-Context Learning,https://arxiv.org/abs/2410.07074,2410.07074,EMNLP,2024,10,arxiv_id,academic, meirtz_context,846,"Components, Techniques and Architectures > Structured Data Integration",NT-LLM: A Novel Node Tokenizer for Integrating Graph Structure into Large Language Models,https://arxiv.org/abs/2410.10743,2410.10743,arXiv,2024,10,arxiv_id,academic, meirtz_context,853,"Components, Techniques and Architectures > Structured Data Integration",CoddLLM: Empowering Large Language Models for Data Analytics,https://arxiv.org/abs/2502.00329,2502.00329,arXiv,2025,2,arxiv_id,academic, meirtz_context,855,"Components, Techniques and Architectures > Structured Data Integration",Structure-Guided Large Language Models for Text-to-SQL Generation,https://arxiv.org/abs/2402.13284,2402.13284,arXiv,2024,2,arxiv_id,academic, meirtz_context,857,"Components, Techniques and Architectures > Structured Data Integration",StructuredRAG: JSON Response Formatting with Large Language Models,https://arxiv.org/abs/2408.11061,2408.11061,arXiv,2024,8,arxiv_id,academic, meirtz_context,866,"Components, Techniques and Architectures > Structured Data Integration",Unifying Large Language Models and Knowledge Graphs: A Roadmap,https://arxiv.org/abs/2306.08302,2306.08302,arXiv,2023,6,arxiv_id,academic, meirtz_context,870,"Components, Techniques and Architectures > Structured Data Integration",Combining Knowledge Graphs and Large Language Models,https://arxiv.org/abs/2407.06564,2407.06564,arXiv,2024,7,arxiv_id,academic, meirtz_context,872,"Components, Techniques and Architectures > Structured Data Integration",All Against Some: Efficient Integration of Large Language Models for Message Passing in Graph Neural Networks,https://arxiv.org/abs/2407.14996,2407.14996,arXiv,2024,7,arxiv_id,academic, meirtz_context,884,"Components, Techniques and Architectures > Self-Generated Context",SelfCite: Self-Supervised Alignment for Context Attribution in Large Language Models,https://arxiv.org/abs/2502.09604,2502.09604,arXiv,2025,2,arxiv_id,academic, meirtz_context,901,"Components, Techniques and Architectures > Self-Generated Context",Self-Consistency Improves Chain of Thought Reasoning in Language Models,https://arxiv.org/abs/2203.11171,2203.11171,ICLR,2022,3,arxiv_id,academic, meirtz_context,903,"Components, Techniques and Architectures > Self-Generated Context",Tree of Thoughts: Deliberate Problem Solving with Large Language Models,https://arxiv.org/abs/2305.10601,2305.10601,arXiv,2023,5,arxiv_id,academic, meirtz_context,907,"Components, Techniques and Architectures > Self-Generated Context",Rethinking Chain-of-Thought from the Perspective of Self-Training,https://arxiv.org/abs/2412.10827,2412.10827,arXiv,2024,12,arxiv_id,academic, meirtz_context,911,"Components, Techniques and Architectures > Self-Generated Context",Autonomous Tree-search Ability of Large Language Models,https://arxiv.org/abs/2310.10686,2310.10686,arXiv,2023,10,arxiv_id,academic, meirtz_context,920,"Components, Techniques and Architectures > Self-Generated Context",Self-Refine: Iterative Refinement with Self-Feedback,https://arxiv.org/abs/2303.17651,2303.17651,arXiv,2023,3,arxiv_id,academic, meirtz_context,924,"Components, Techniques and Architectures > Self-Generated Context","Reflect, Retry, Reward: Self-Improving LLMs via Reinforcement Learning",https://arxiv.org/abs/2505.24726,2505.24726,arXiv,2025,5,arxiv_id,academic, meirtz_context,926,"Components, Techniques and Architectures > Self-Generated Context",Large Language Models Can Self-Improve in Long-context Reasoning,https://arxiv.org/abs/2411.08147,2411.08147,arXiv,2024,11,arxiv_id,academic, meirtz_context,930,"Components, Techniques and Architectures > Self-Generated Context",Code Generation with AlphaCodium: From Prompt Engineering to Flow Engineering,https://arxiv.org/abs/2401.08500,2401.08500,arXiv,2024,1,arxiv_id,academic, meirtz_context,932,"Components, Techniques and Architectures > Self-Generated Context",Language Agent Tree Search Unifies Reasoning Acting and Planning in Language Models,https://arxiv.org/abs/2310.04406,2310.04406,arXiv,2023,10,arxiv_id,academic, meirtz_context,941,"Components, Techniques and Architectures > Self-Generated Context",EvoPrompt: Connecting LLMs with Evolutionary Algorithms Yields Powerful Prompt Optimizers,https://arxiv.org/abs/2309.08532,2309.08532,ICLR,2023,9,arxiv_id,academic, meirtz_context,945,"Components, Techniques and Architectures > Self-Generated Context",AutoPDL: Automatic Prompt Optimization for LLM Agents,https://arxiv.org/abs/2504.04365,2504.04365,AutoML,2025,4,arxiv_id,academic, meirtz_context,947,"Components, Techniques and Architectures > Self-Generated Context",Agent-Pro: Learning to Evolve Coder Agents via Proposal-based Programming,https://arxiv.org/abs/2402.17574,2402.17574,arXiv,2024,2,arxiv_id,academic, meirtz_context,969,Implementation and Challenges > 0. Agent Harnesses and Runtime Systems,Building Effective Agents,https://www.anthropic.com/engineering/building-effective-agents,,Anthropic,2024,12,badge,grey, meirtz_context,970,Implementation and Challenges > 0. Agent Harnesses and Runtime Systems,OpenAI Agents Guide,https://platform.openai.com/docs/guides/agents,,OpenAI,,,,grey, meirtz_context,971,Implementation and Challenges > 0. Agent Harnesses and Runtime Systems,Google Agent Development Kit (ADK),https://google.github.io/adk-docs/,,Google,,,,grey, meirtz_context,972,Implementation and Challenges > 0. Agent Harnesses and Runtime Systems,LangChain Deep Agents Overview,https://docs.langchain.com/oss/python/deepagents/overview,,LangChain,,,,grey, meirtz_context,973,Implementation and Challenges > 0. Agent Harnesses and Runtime Systems,Microsoft Agent Framework Overview,https://learn.microsoft.com/en-us/agent-framework/user-guide/overview,,Microsoft,,,,grey, meirtz_context,974,Implementation and Challenges > 0. Agent Harnesses and Runtime Systems,Bring Your AI,https://bringyour.ai/claude-code-to-codex,,Tool,,,,grey, meirtz_context,991,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Retrieval-Augmented Generation for Large Language Models: A Survey,https://arxiv.org/abs/2312.10997,2312.10997,arXiv,2023,12,arxiv_id,academic, meirtz_context,996,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),A Survey of Graph Retrieval-Augmented Generation for Customized Large Language Models,https://arxiv.org/abs/2501.13958,2501.13958,arXiv,2025,1,arxiv_id,academic, meirtz_context,1001,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Retrieval Augmented Generation (RAG) and Beyond: A Comprehensive Survey on How to Make your LLMs use External Data More Wisely,https://arxiv.org/abs/2409.14924,2409.14924,arXiv,2024,9,arxiv_id,academic, meirtz_context,1003,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Evaluation of Retrieval-Augmented Generation: A Survey,https://arxiv.org/abs/2405.07437,2405.07437,arXiv,2024,5,arxiv_id,academic, meirtz_context,1008,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks,https://arxiv.org/abs/2005.11401,2005.11401,arXiv,2020,5,arxiv_id,academic, meirtz_context,1012,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),A Survey on Knowledge-Oriented Retrieval-Augmented Generation,https://arxiv.org/abs/2503.10677,2503.10677,arXiv,2025,3,arxiv_id,academic, meirtz_context,1016,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),A Survey on RAG Meeting LLMs: Towards Retrieval-Augmented Large Language Models,https://arxiv.org/abs/2405.06211,2405.06211,arXiv,2024,5,arxiv_id,academic, meirtz_context,1025,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Beyond the Limits: A Survey of Techniques to Extend the Context Length in Large Language Models,https://arxiv.org/abs/2402.02244,2402.02244,arXiv,2024,2,arxiv_id,academic, meirtz_context,1027,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),In-context Examples Selection for Machine Translation,https://arxiv.org/abs/2212.02437,2212.02437,arXiv,2022,12,arxiv_id,academic, meirtz_context,1029,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),In Defense of RAG in the Era of Long-Context Language Models,https://arxiv.org/abs/2409.01666,2409.01666,arXiv,2024,9,arxiv_id,academic, meirtz_context,1031,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks,https://arxiv.org/abs/2005.11401,2005.11401,arXiv,2020,5,arxiv_id,academic, meirtz_context,1033,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),LightRAG: Simple and Fast Retrieval-Augmented Generation,https://arxiv.org/abs/2410.05779,2410.05779,arXiv,2024,10,arxiv_id,academic, meirtz_context,1038,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Generate rather than Retrieve: Large Language Models are Strong Context Generators,https://arxiv.org/abs/2209.10063,2209.10063,arXiv,2022,9,arxiv_id,academic, meirtz_context,1043,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Large language models can be easily distracted by irrelevant context,https://arxiv.org/abs/2302.00093,2302.00093,arXiv,2023,2,arxiv_id,academic, meirtz_context,1050,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Dense Passage Retrieval for Open-Domain Question Answering,https://arxiv.org/abs/2004.04906,2004.04906,arXiv,2020,4,arxiv_id,academic, meirtz_context,1062,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models through Question Complexity,https://arxiv.org/abs/2403.14403,2403.14403,arXiv,2024,3,arxiv_id,academic, meirtz_context,1067,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Improving language models by retrieving from trillions of tokens,https://arxiv.org/abs/2112.04426,2112.04426,arXiv,2021,12,arxiv_id,academic, meirtz_context,1071,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),IM-RAG: Multi-Round Retrieval-Augmented Generation Through Learning Inner Monologues,https://arxiv.org/abs/2405.13021,2405.13021,arXiv,2024,5,arxiv_id,academic, meirtz_context,1073,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),RAGCache: Efficient Knowledge Caching for Retrieval-Augmented Generation,https://arxiv.org/abs/2404.12457,2404.12457,arXiv,2024,4,arxiv_id,academic, meirtz_context,1075,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Corrective Retrieval Augmented Generation,https://arxiv.org/abs/2401.15884,2401.15884,arXiv,2024,1,arxiv_id,academic, meirtz_context,1080,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),RankRAG: Unifying Context Ranking with Retrieval-Augmented Generation in LLMs,https://arxiv.org/abs/2407.02485,2407.02485,arXiv,2024,7,arxiv_id,academic, meirtz_context,1082,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Astute RAG: Overcoming Imperfect Retrieval Augmentation and Knowledge Conflicts for Large Language Models,https://arxiv.org/abs/2410.07176,2410.07176,arXiv,2024,10,arxiv_id,academic, meirtz_context,1084,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Learning to Filter Context for Retrieval-Augmented Generation,https://arxiv.org/abs/2311.08377,2311.08377,arXiv,2023,11,arxiv_id,academic, meirtz_context,1089,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Query Rewriting in Retrieval-Augmented Large Language Models,https://arxiv.org/abs/2305.14283,2305.14283,arXiv,2023,5,arxiv_id,academic, meirtz_context,1094,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),UPRISE: Universal Prompt Retrieval for Improving Zero-Shot Evaluation,https://arxiv.org/abs/2303.08518,2303.08518,arXiv,2023,3,arxiv_id,academic, meirtz_context,1099,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Longllmlingua: Accelerating and enhancing llms in long context scenarios via prompt compression,https://arxiv.org/abs/2310.06839,2310.06839,arXiv,2023,10,arxiv_id,academic, meirtz_context,1104,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Document-level event argument extraction by conditional generation,https://arxiv.org/abs/2104.05919,2104.05919,arXiv,2021,4,arxiv_id,academic, meirtz_context,1109,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Multi-sentence Argument Linking,https://arxiv.org/abs/1911.03766,1911.03766,arXiv,2019,11,arxiv_id,academic, meirtz_context,1114,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Fine-tuning or retrieval? comparing knowledge injection in llms,https://arxiv.org/abs/2312.05934,2312.05934,arXiv,2023,12,arxiv_id,academic, meirtz_context,1116,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),IAG: Induction-Augmented Generation Framework for Answering Reasoning Questions,https://arxiv.org/abs/2311.18397,2311.18397,arXiv,2023,11,arxiv_id,academic, meirtz_context,1118,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Retrieval Meets Long Context Large Language Models,https://arxiv.org/abs/2310.03025,2310.03025,arXiv,2023,10,arxiv_id,academic, meirtz_context,1120,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Dense x retrieval: What retrieval granularity should we use?,https://arxiv.org/abs/2312.06648,2312.06648,arXiv,2023,12,arxiv_id,academic, meirtz_context,1125,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Investigating the Factual Knowledge Boundary of Large Language Models with Retrieval Augmentation,https://arxiv.org/abs/2307.11019,2307.11019,arXiv,2023,7,arxiv_id,academic, meirtz_context,1130,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),The Power of Noise: Redefining Retrieval for RAG Systems,https://arxiv.org/abs/2401.14887,2401.14887,arXiv,2024,1,arxiv_id,academic, meirtz_context,1135,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),RECITATION-AUGMENTED LANGUAGE MODELS,https://arxiv.org/abs/2210.01296,2210.01296,arXiv,2022,10,arxiv_id,academic, meirtz_context,1140,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Robust Retrieval Augmented Generation for Zero-shot Slot Filling,https://arxiv.org/abs/2108.13934,2108.13934,arXiv,2021,8,arxiv_id,academic, meirtz_context,1145,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),In-Context Retrieval-Augmented Language Models,https://arxiv.org/abs/2302.00083,2302.00083,arXiv,2023,2,arxiv_id,academic, meirtz_context,1150,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Learning to Retrieve In-Context Examples for Large Language Models,https://arxiv.org/abs/2307.07164,2307.07164,arXiv,2023,7,arxiv_id,academic, meirtz_context,1162,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),FlashRAG: A Modular Toolkit for Efficient Retrieval-Augmented Generation Research,https://arxiv.org/abs/2405.13576,2405.13576,arXiv,2024,5,arxiv_id,academic, meirtz_context,1167,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Multi-Head RAG: Solving Multi-Aspect Problems with LLMs,https://arxiv.org/abs/2406.05085,2406.05085,arXiv,2024,6,arxiv_id,academic, meirtz_context,1172,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),StructRAG: Boosting Knowledge Intensive Reasoning of LLMs via Inference-time Hybrid Information Structurization,https://arxiv.org/abs/2410.08815,2410.08815,arXiv,2024,10,arxiv_id,academic, meirtz_context,1177,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),RAFT: Adapting Language Model to Domain Specific RAG,https://arxiv.org/abs/2403.10131,2403.10131,arXiv,2024,3,arxiv_id,academic, meirtz_context,1182,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Retrieval-Generation Alignment for End-to-End Task-Oriented Dialogue System,https://arxiv.org/abs/2310.08877,2310.08877,arXiv,2023,10,arxiv_id,academic, meirtz_context,1187,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),UniMS-RAG: A Unified Multi-source Retrieval-Augmented Generation for Personalized Dialogue Systems,https://arxiv.org/abs/2401.13256,2401.13256,arXiv,2024,1,arxiv_id,academic, meirtz_context,1191,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),RA-DIT: RETRIEVAL-AUGMENTED DUAL INSTRUCTION TUNING,https://arxiv.org/abs/2310.01352,2310.01352,arXiv,2023,10,arxiv_id,academic, meirtz_context,1196,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Self-Knowledge Guided Retrieval Augmentation for Large Language Models,https://arxiv.org/abs/2310.05002,2310.05002,arXiv,2023,10,arxiv_id,academic, meirtz_context,1201,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Prompt-Guided Retrieval Augmentation for Non-Knowledge-Intensive Tasks,https://arxiv.org/abs/2305.17653,2305.17653,arXiv,2023,5,arxiv_id,academic, meirtz_context,1206,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),REPLUG: Retrieval-Augmented Black-Box Language Models,https://arxiv.org/abs/2301.12652,2301.12652,arXiv,2023,1,arxiv_id,academic, meirtz_context,1208,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Query Rewriting for Retrieval-Augmented Large Language Models,https://doi.org/10.18653/v1/2023.emnlp-main.323,,EMNLP,2023,,badge,academic, meirtz_context,1213,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Lift Yourself Up: Retrieval-augmented Text Generation with Self-Memory,https://arxiv.org/abs/2305.02437,2305.02437,arXiv,2023,5,arxiv_id,academic, meirtz_context,1218,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Improving the Domain Adaptation of Retrieval Augmented Generation (RAG) Models for Open Domain Question Answering,https://arxiv.org/abs/2210.02627,2210.02627,arXiv,2022,10,arxiv_id,academic, meirtz_context,1227,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Don't Forget to Connect! Improving RAG with Graph-based Reranking,https://arxiv.org/abs/2405.18414,2405.18414,arXiv,2024,5,arxiv_id,academic, meirtz_context,1229,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),From Local to Global: A Graph RAG Approach to Query-Focused Summarization,https://arxiv.org/abs/2404.16130,2404.16130,arXiv,2024,4,arxiv_id,academic, meirtz_context,1231,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),GRAG: Graph Retrieval-Augmented Generation,https://arxiv.org/abs/2405.16506,2405.16506,arXiv,2024,5,arxiv_id,academic, meirtz_context,1236,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Iseeq: Information seeking question generation using dynamic meta-information retrieval and knowledge graphs,https://arxiv.org/abs/2112.07622,2112.07622,arXiv,2021,12,arxiv_id,academic, meirtz_context,1241,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),G-retriever: Retrieval-augmented generation for textual graph understanding and question answering,https://arxiv.org/abs/2402.07630,2402.07630,arXiv,2024,2,arxiv_id,academic, meirtz_context,1246,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Knowledge graph prompting for multi-document question answering,https://arxiv.org/abs/2402.08774,2402.08774,arXiv,2024,2,arxiv_id,academic, meirtz_context,1251,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),GNN-RAG: Graph Neural Retrieval for Large Language Model Reasoning,https://arxiv.org/abs/2405.20139,2405.20139,arXiv,2024,5,arxiv_id,academic, meirtz_context,1276,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Mitigating Large Language Model Hallucinations via Autonomous Knowledge Graph-based Retrofitting,https://arxiv.org/abs/2311.13314,2311.13314,arXiv,2023,11,arxiv_id,academic, meirtz_context,1281,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),In-depth Analysis of Graph-based RAG in a Unified Framework,https://arxiv.org/abs/2503.04338,2503.04338,arXiv,2025,3,arxiv_id,academic, meirtz_context,1286,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval,https://arxiv.org/abs/2401.18059,2401.18059,arXiv,2024,1,arxiv_id,academic, meirtz_context,1291,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),TableRAG: Million-Token Table Understanding with Language Models,https://arxiv.org/abs/2410.04739,2410.04739,arXiv,2024,10,arxiv_id,academic, meirtz_context,1296,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),KAG: Boosting LLMs in Professional Domains via Knowledge Augmented Generation,https://arxiv.org/abs/2409.13731,2409.13731,arXiv,2024,9,arxiv_id,academic, meirtz_context,1301,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),GFM-RAG: Graph Foundation Model for Retrieval Augmented Generation,https://arxiv.org/abs/2502.01113,2502.01113,arXiv,2025,2,arxiv_id,academic, meirtz_context,1316,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),From RAG to Memory: Non-Parametric Continual Learning for Large Language Models,https://arxiv.org/abs/2502.14802,2502.14802,arXiv,2025,2,arxiv_id,academic, meirtz_context,1321,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models,https://arxiv.org/abs/2405.14924,2405.14924,arXiv,2024,5,arxiv_id,academic, meirtz_context,1326,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),GraphReader: Building Graph-based Agent to Enhance Long-Context Abilities of Large Language Models,https://arxiv.org/abs/2406.14550,2406.14550,arXiv,2024,6,arxiv_id,academic, meirtz_context,1328,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),PlanRAG: A Plan-then-Retrieval Augmented Generation for Generative Large Language Models as Decision Makers,https://arxiv.org/abs/2406.12430,2406.12430,arXiv,2024,6,arxiv_id,academic, meirtz_context,1333,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),"Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection",https://arxiv.org/abs/2402.08353,2402.08353,arXiv,2024,2,arxiv_id,academic, meirtz_context,1338,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),DeepRAG: Thinking to Retrieve Step by Step for Large Language Models,https://arxiv.org/abs/2502.01142,2502.01142,arXiv,2025,2,arxiv_id,academic, meirtz_context,1340,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Paperqa: Retrieval-augmented generative agent for scientific research,https://arxiv.org/abs/2312.07559,2312.07559,arXiv,2023,12,arxiv_id,academic, meirtz_context,1342,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Large Language Models as Source Planner for Personalized Knowledge-grounded Dialogues,https://arxiv.org/abs/2308.06181,2308.06181,arXiv,2023,8,arxiv_id,academic, meirtz_context,1347,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),PRCA: Fitting Black-Box Large Language Models for Retrieval Question Answering via Pluggable Reward-Driven Contextual Adapter,https://arxiv.org/abs/2310.18347,2310.18347,arXiv,2023,10,arxiv_id,academic, meirtz_context,1352,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),"SELF-RAG: LEARNING TO RETRIEVE, GENERATE, AND CRITIQUE THROUGH SELF-REFLECTION",https://arxiv.org/abs/2310.11511,2310.11511,arXiv,2023,10,arxiv_id,academic, meirtz_context,1357,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Horizon Generation,https://arxiv.org/abs/2403.05313,2403.05313,arXiv,2024,3,arxiv_id,academic, meirtz_context,1362,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Chain-of-verification reduces hallucination in large language models,https://arxiv.org/abs/2309.11495,2309.11495,arXiv,2023,9,arxiv_id,academic, meirtz_context,1364,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),HM-RAG: Hierarchical Multi-Agent Multimodal Retrieval Augmented Generation,https://arxiv.org/abs/2504.12330,2504.12330,arXiv,2025,4,arxiv_id,academic, meirtz_context,1368,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries,https://arxiv.org/abs/2401.15391,2401.15391,arXiv,2024,1,arxiv_id,academic, meirtz_context,1372,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),MMOA-RAG: Improving Retrieval-Augmented Generation through Multi-Agent Reinforcement Learning,https://arxiv.org/abs/2010.10110,2010.10110,arXiv,2020,10,arxiv_id,academic, meirtz_context,1376,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),"Search-in-the-Chain: Towards Accurate, Credible, and Up-to-Date Large Language Models",https://arxiv.org/abs/2304.14732,2304.14732,arXiv,2023,4,arxiv_id,academic, meirtz_context,1384,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),StreamingRAG: Real-time Contextual Retrieval and Generation Framework,https://arxiv.org/abs/2501.14101,2501.14101,arXiv,2025,1,arxiv_id,academic, meirtz_context,1388,Implementation and Challenges > 1. Retrieval-Augmented Generation (RAG),Multi-task Retriever Fine-tuning for Domain-Specific and Efficient RAG,https://arxiv.org/abs/2501.04652,2501.04652,arXiv,2025,1,arxiv_id,academic, meirtz_context,1406,Implementation and Challenges > 2. Memory Systems > Runtime Memory Design Patterns,LangGraph Memory Overview,https://docs.langchain.com/oss/javascript/langgraph/memory,,LangChain,,,,grey, meirtz_context,1407,Implementation and Challenges > 2. Memory Systems > Runtime Memory Design Patterns,Letta Memory Blocks,https://docs.letta.com/guides/core-concepts/memory/memory-blocks,,Letta,,,,grey, meirtz_context,1408,Implementation and Challenges > 2. Memory Systems > Runtime Memory Design Patterns,Claude Code Memory,https://docs.anthropic.com/en/docs/claude-code/memory,,Anthropic,,,,grey, meirtz_context,1417,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Introducing Codex,https://openai.com/index/introducing-codex/,,OpenAI,2025,5,badge,grey, meirtz_context,1418,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Claude Code Memory,https://docs.anthropic.com/en/docs/claude-code/memory,,Anthropic,,,,grey, meirtz_context,1419,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Claude Code Subagents,https://docs.anthropic.com/en/docs/claude-code/sub-agents,,Anthropic,,,,grey, meirtz_context,1420,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,LangChain Deep Agents Overview,https://docs.langchain.com/oss/python/deepagents/overview,,LangChain,,,,grey, meirtz_context,1425,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,MemGPT: Towards LLMs as Operating Systems,https://arxiv.org/abs/2310.08560,2310.08560,arXiv,2023,10,arxiv_id,academic, meirtz_context,1429,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory,https://arxiv.org/abs/2504.19413,2504.19413,arXiv,2025,4,arxiv_id,academic, meirtz_context,1433,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,MemoryLLM: Towards Self-Updatable Large Language Models,https://arxiv.org/abs/2402.04624,2402.04624,arXiv,2024,2,arxiv_id,academic, meirtz_context,1437,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Infinite-LLM: Efficient LLM Service for Long Context with DistAttention and Distributed KVCache,https://arxiv.org/abs/2401.02669,2401.02669,arXiv,2024,1,arxiv_id,academic, meirtz_context,1439,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Memory-Augmented Generative Adversarial Transformers,https://arxiv.org/abs/2402.19218,2402.19218,arXiv,2024,2,arxiv_id,academic, meirtz_context,1445,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,PAM (Portable AI Memory): An Open Interchange Format for AI User Memories,https://portable-ai-memory.org/spec/v1.0/,,,,,,grey, meirtz_context,1453,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Survey on Memory-Augmented Neural Networks: Cognitive Insights to AI Applications,https://arxiv.org/abs/2312.06141,2312.06141,arXiv,2023,12,arxiv_id,academic, meirtz_context,1455,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,"A Machine with Short-Term, Episodic, and Semantic Memory Systems",https://arxiv.org/abs/2212.02098,2212.02098,arXiv,2022,12,arxiv_id,academic, meirtz_context,1459,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,From Human Memory to AI Memory: A Survey on Memory Mechanisms in the Era of LLMs,https://arxiv.org/abs/2504.15965,2504.15965,arXiv,2025,4,arxiv_id,academic, meirtz_context,1466,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,The Role of Memory in LLMs: Persistent Context for Smarter Conversations,https://doi.org/10.18535/ijsrm/v12i11.ec04,,IJSRM,2024,11,badge,academic, meirtz_context,1468,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Episodic Memory in AI Agents Poses Risks that Should Be Studied and Mitigated,https://arxiv.org/abs/2401.11739,2401.11739,arXiv,2024,1,arxiv_id,academic, meirtz_context,1470,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Larimar: Large Language Models with Episodic Memory Control,https://arxiv.org/abs/2403.11901,2403.11901,ICML,2024,3,arxiv_id,academic, meirtz_context,1472,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,EM-LLM: Human-like Episodic Memory for Infinite Context LLMs,https://arxiv.org/abs/2407.09450,2407.09450,ICLR,2024,7,arxiv_id,academic, meirtz_context,1476,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Large Language Models with Controllable Working Memory,https://arxiv.org/abs/2211.05110,2211.05110,arXiv,2022,11,arxiv_id,academic, meirtz_context,1478,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Empowering Working Memory for Large Language Model Agents,https://arxiv.org/abs/2312.17259,2312.17259,arXiv,2023,12,arxiv_id,academic, meirtz_context,1494,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,MemoChat: Tuning LLMs to Use Memos for Consistent Long-Range Open-Domain Conversation,https://arxiv.org/abs/2308.08239,2308.08239,arXiv,2023,8,arxiv_id,academic, meirtz_context,1496,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Think-in-Memory: Recalling and Post-thinking Enable LLMs with Long-Term Memory,https://arxiv.org/abs/2311.08719,2311.08719,arXiv,2023,11,arxiv_id,academic, meirtz_context,1498,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Generative Agents: Interactive Simulacra of Human Behavior,https://arxiv.org/abs/2304.03442,2304.03442,arXiv,2023,4,arxiv_id,academic, meirtz_context,1500,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Self-Controlled Memory Framework for Large Language Models,https://arxiv.org/abs/2304.13343,2304.13343,arXiv,2023,4,arxiv_id,academic, meirtz_context,1506,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Personalized LLM Response Generation with Parameterized User Memory Injection,https://arxiv.org/abs/2404.03565,2404.03565,arXiv,2024,4,arxiv_id,academic, meirtz_context,1508,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Soul-Driven Interaction Design: A Position Paper on Declarative Persona Specifications for AI Agents,https://doi.org/10.5281/zenodo.18678616,,Zenodo,2026,2,badge,academic, meirtz_context,1510,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Soul Spec — Open Specification for AI Agent Persona Packages,https://clawsouls.ai/spec,,,,,,grey, meirtz_context,1518,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Constitutional AI: Harmlessness from AI Feedback,https://arxiv.org/abs/2212.08073,2212.08073,arXiv,2022,12,arxiv_id,academic, meirtz_context,1520,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Improving alignment of dialogue agents via targeted human judgements (Sparrow),https://arxiv.org/abs/2209.14375,2209.14375,arXiv,2022,9,arxiv_id,academic, meirtz_context,1526,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,WebGPT: Browser-assisted question-answering with human feedback,https://arxiv.org/abs/2112.09332,2112.09332,arXiv,2021,12,arxiv_id,academic, meirtz_context,1528,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs,https://arxiv.org/abs/2307.16789,2307.16789,arXiv,2023,7,arxiv_id,academic, meirtz_context,1534,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Language Models are Few-Shot Learners (GPT-3),https://arxiv.org/abs/2005.14165,2005.14165,arXiv,2020,5,arxiv_id,academic, meirtz_context,1536,Implementation and Challenges > 2. Memory Systems > Project Memory and Instruction Artifacts,Reflexion: Language Agents with Verbal Reinforcement Learning,https://arxiv.org/abs/2303.11366,2303.11366,NeurIPS,2023,3,arxiv_id,academic, meirtz_context,1548,Implementation and Challenges > 3. Agent Communication,A Survey of AI Agent Protocols,https://arxiv.org/abs/2504.16736,2504.16736,arXiv,2025,4,arxiv_id,academic, meirtz_context,1553,Implementation and Challenges > 3. Agent Communication,A Survey of Multi-Agent Deep Reinforcement Learning with Communication,https://arxiv.org/abs/2203.08975,2203.08975,arXiv,2022,3,arxiv_id,academic, meirtz_context,1555,Implementation and Challenges > 3. Agent Communication,Beyond Self-Talk: A Communication-Centric Survey of LLM-Based Multi-Agent Systems,https://arxiv.org/abs/2502.14321,2502.14321,arXiv,2025,2,arxiv_id,academic, meirtz_context,1557,Implementation and Challenges > 3. Agent Communication,Large Language Model based Multi-Agents: A Survey of Progress and Challenges,https://arxiv.org/abs/2402.01680,2402.01680,arXiv,2024,2,arxiv_id,academic, meirtz_context,1577,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,Model Context Protocol Specification,https://modelcontextprotocol.io/specification/2025-06-18,,Spec,2025,6,badge,grey, meirtz_context,1578,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,Model Context Protocol Architecture,https://modelcontextprotocol.io/docs/learn/architecture,,Docs,,,,grey, meirtz_context,1579,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,Agent2Agent Protocol (A2A),https://a2a-protocol.org/latest/,,Protocol,,,,grey, meirtz_context,1580,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,AG-UI Documentation,https://docs.ag-ui.com/,,Protocol,,,,grey, meirtz_context,1581,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,ACP Connect,https://docs.agntcy.org/syntactic/connect/,,Protocol,,,,grey, meirtz_context,1582,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,AgentSchema,https://microsoft.github.io/AgentSchema/,,Schema,,,,grey, meirtz_context,1588,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,"A survey of agent interoperability protocols: Model Context Protocol (MCP), Agent Communication Protocol (ACP), and Agent-to-Agent Protocol (A2A)",https://arxiv.org/abs/2505.02279,2505.02279,arXiv,2025,5,arxiv_id,academic, meirtz_context,1590,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,Expressive Multi-Agent Communication via Identity-Aware Learning,https://doi.org/10.1609/aaai.v38i16.29683,,AAAI,2024,3,badge,academic, meirtz_context,1592,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,Context-aware Communication for Multi-agent Reinforcement Learning (CACOM),https://arxiv.org/abs/2312.15600,2312.15600,arXiv,2023,12,arxiv_id,academic, meirtz_context,1596,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,"A survey of agent interoperability protocols: Model Context Protocol (MCP), Agent Communication Protocol (ACP), Agent-to-Agent Protocol (A2A), and Agent Network Protocol (ANP)",https://arxiv.org/abs/2505.02279,2505.02279,arXiv,2025,5,arxiv_id,academic, meirtz_context,1598,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,Agent Capability Negotiation and Binding Protocol (ACNBP),https://arxiv.org/abs/2506.13590,2506.13590,arXiv,2025,6,arxiv_id,academic, meirtz_context,1600,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,A Scalable Communication Protocol for Networks of Large Language Models,https://arxiv.org/abs/2410.11905,2410.11905,arXiv,2024,10,arxiv_id,academic, meirtz_context,1620,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,Not Human Search,https://nothumansearch.ai/,,Tool,,,,grey, meirtz_context,1634,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,Task-Agnostic Contrastive Pre-Training for Inter-Agent Communication,https://arxiv.org/abs/2501.02174,2501.02174,AAMAS,2025,1,arxiv_id,academic, meirtz_context,1636,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,AC2C: Adaptively Controlled Two-Hop Communication for Multi-Agent Reinforcement Learning,https://arxiv.org/abs/2302.12515,2302.12515,arXiv,2023,2,arxiv_id,academic, meirtz_context,1638,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,"CAMEL: Communicative Agents for ""Mind"" Exploration of Large Language Model Society",https://arxiv.org/abs/2303.17760,2303.17760,arXiv,2023,3,arxiv_id,academic, meirtz_context,1643,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,Context-aware Communication for Multi-agent Reinforcement Learning (CACOM),https://arxiv.org/abs/2312.15600,2312.15600,arXiv,2023,12,arxiv_id,academic, meirtz_context,1650,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,Expressive Multi-Agent Communication via Identity-Aware Learning,https://arxiv.org/abs/2401.07872,2401.07872,arXiv,2024,1,arxiv_id,academic, meirtz_context,1652,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,MAGIS: LLM-Based Multi-Agent Framework for GitHub Issue Resolution,https://arxiv.org/abs/2403.17927,2403.17927,arXiv,2024,3,arxiv_id,academic, meirtz_context,1654,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,AutoAgents: A Framework for Automatic Agent Generation,https://arxiv.org/abs/2309.17288,2309.17288,arXiv,2023,9,arxiv_id,academic, meirtz_context,1659,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,MDTeamGPT: A Self-Evolving LLM-based Multi-Agent Framework for Multi-Disciplinary Team Medical Consultation,https://arxiv.org/abs/2503.13856,2503.13856,arXiv,2025,3,arxiv_id,academic, meirtz_context,1664,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation,https://arxiv.org/abs/2308.08155,2308.08155,arXiv,2023,8,arxiv_id,academic, meirtz_context,1673,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,ProAgent: Building Proactive Cooperative AI with Large Language Models,https://arxiv.org/abs/2308.11339,2308.11339,arXiv,2023,8,arxiv_id,academic, meirtz_context,1678,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,Improving Factuality and Reasoning in Language Models through Multiagent Debate,https://arxiv.org/abs/2305.14325,2305.14325,arXiv,2023,5,arxiv_id,academic, meirtz_context,1683,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,ChatDev: Communicative Agents for Software Development,https://arxiv.org/abs/2307.07924,2307.07924,arXiv,2023,7,arxiv_id,academic, meirtz_context,1688,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,Multi-Agent Incentive Communication via Decentralized Teammate Modeling,https://arxiv.org/abs/2310.10436,2310.10436,arXiv,2023,10,arxiv_id,academic, meirtz_context,1693,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,AgentCoord: Visually Exploring Coordination Strategy for LLM-based Multi-Agent Collaboration,https://arxiv.org/abs/2404.11943,2404.11943,arXiv,2024,4,arxiv_id,academic, meirtz_context,1698,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,Simulating Opinion Dynamics with Networks of LLM-based Agents,https://arxiv.org/abs/2311.09618,2311.09618,arXiv,2023,11,arxiv_id,academic, meirtz_context,1703,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework,https://arxiv.org/abs/2308.00352,2308.00352,arXiv,2023,8,arxiv_id,academic, meirtz_context,1708,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,Chain of Agents: Large Language Models Collaborating on Long-Context Tasks,https://arxiv.org/abs/2406.02818,2406.02818,arXiv,2024,6,arxiv_id,academic, meirtz_context,1710,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,Multi-Agent Incentive Communication via Decentralized Teammate Modeling,https://doi.org/10.1609/aaai.v36i9.21179,,,,,,academic, meirtz_context,1713,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,ProAgent: Building Proactive Cooperative Agents with Large Language Models,https://doi.org/10.1609/aaai.v38i16.29710,,AAAI,2024,3,badge,academic, meirtz_context,1717,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,Model Context Protocol (MCP),https://github.com/modelcontextprotocol,,GitHub,,,,grey, meirtz_context,1719,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,CoMAS: Co-Evolving Multi-Agent Systems via Interaction Rewards,https://arxiv.org/abs/2510.08529,2510.08529,arXiv,2025,10,arxiv_id,academic, meirtz_context,1721,Implementation and Challenges > 3. Agent Communication > Open Agent Protocols and Interoperability,Achilles Heel of Distributed Multi-Agent Systems,https://arxiv.org/abs/2504.07461,2504.07461,arXiv,2025,4,arxiv_id,academic, meirtz_context,1735,Implementation and Challenges > 4. Tool Use and Function Calling > Hosted Agent Tools and Computer Use,OpenAI Tools Guide,https://developers.openai.com/api/docs/guides/tools,,OpenAI,,,,grey, meirtz_context,1736,Implementation and Challenges > 4. Tool Use and Function Calling > Hosted Agent Tools and Computer Use,Introducing Codex,https://openai.com/index/introducing-codex/,,OpenAI,2025,5,badge,grey, meirtz_context,1737,Implementation and Challenges > 4. Tool Use and Function Calling > Hosted Agent Tools and Computer Use,Computer Use for Claude 3.5,https://www.anthropic.com/news/3-5-models-and-computer-use,,Anthropic,2024,10,badge,grey, meirtz_context,1738,Implementation and Challenges > 4. Tool Use and Function Calling > Hosted Agent Tools and Computer Use,Google Vertex AI Agent Engine,https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/overview,,Google,,,,grey, meirtz_context,1739,Implementation and Challenges > 4. Tool Use and Function Calling > Hosted Agent Tools and Computer Use,OSWorld,https://os-world.github.io/,,Benchmark,,,,grey, meirtz_context,1740,Implementation and Challenges > 4. Tool Use and Function Calling > Hosted Agent Tools and Computer Use,Lumen,https://github.com/omxyz/lumen,,,,,,repository, meirtz_context,1745,Implementation and Challenges > 4. Tool Use and Function Calling > Hosted Agent Tools and Computer Use,Toolformer: Language Models Can Teach Themselves to Use Tools,https://arxiv.org/abs/2302.04761,2302.04761,NeurIPS,2023,2,arxiv_id,academic, meirtz_context,1749,Implementation and Challenges > 4. Tool Use and Function Calling > Hosted Agent Tools and Computer Use,ReAct: Synergizing Reasoning and Acting in Language Models,https://arxiv.org/abs/2210.03629,2210.03629,arXiv,2022,10,arxiv_id,academic, meirtz_context,1753,Implementation and Challenges > 4. Tool Use and Function Calling > Hosted Agent Tools and Computer Use,Augmented Language Models: a Survey,https://arxiv.org/abs/2302.07842,2302.07842,arXiv,2023,2,arxiv_id,academic, meirtz_context,1755,Implementation and Challenges > 4. Tool Use and Function Calling > Hosted Agent Tools and Computer Use,Tool Learning with Large Language Models: A Survey,https://arxiv.org/abs/2405.17935,2405.17935,arXiv,2024,5,arxiv_id,academic, meirtz_context,1764,Implementation and Challenges > 4. Tool Use and Function Calling > Hosted Agent Tools and Computer Use,Granite-Function Calling Model: Introducing Function Calling Abilities via Multi-task Learning of Granular Tasks,https://arxiv.org/abs/2407.00121,2407.00121,arXiv,2024,7,arxiv_id,academic, meirtz_context,1766,Implementation and Challenges > 4. Tool Use and Function Calling > Hosted Agent Tools and Computer Use,HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face,https://arxiv.org/abs/2303.17580,2303.17580,NeurIPS,2023,3,arxiv_id,academic, meirtz_context,1770,Implementation and Challenges > 4. Tool Use and Function Calling > Hosted Agent Tools and Computer Use,"Enhancing Function-Calling Capabilities in LLMs: Strategies for Prompt Formats, Data Integration, and Multilingual Translation",https://arxiv.org/abs/2412.01130,2412.01130,NAACL,2024,12,arxiv_id,academic, meirtz_context,1772,Implementation and Challenges > 4. Tool Use and Function Calling > Hosted Agent Tools and Computer Use,A Real-World WebAgent for Complex Web Tasks,https://arxiv.org/html/2307.12856v4,2307.12856,arXiv,2023,7,arxiv_id,academic, meirtz_context,1795,Evaluation Paradigms for Context-Driven Systems > Context Quality Assessment,RULER: What's the Real Context Size of Your Long-Context Language Models?,https://arxiv.org/abs/2404.06654,2404.06654,COLM,2024,4,arxiv_id,academic, meirtz_context,1800,Evaluation Paradigms for Context-Driven Systems > Context Quality Assessment,∞BENCH: Extending Long Context Evaluation Beyond 100K Tokens,https://arxiv.org/abs/2402.13718,2402.13718,ACL,2024,2,arxiv_id,academic, meirtz_context,1821,Evaluation Paradigms for Context-Driven Systems > Context Quality Assessment,BBScore: A Brownian Bridge Based Metric for Assessing Text Coherence,https://doi.org/10.1609/aaai.v38i13.29414,,AAAI,2024,3,badge,academic, meirtz_context,1830,Evaluation Paradigms for Context-Driven Systems > Context Quality Assessment,Evaluation of Retrieval-Augmented Generation: A Survey,https://arxiv.org/abs/2405.07437,2405.07437,arXiv,2024,5,arxiv_id,academic, meirtz_context,1831,Evaluation Paradigms for Context-Driven Systems > Context Quality Assessment,Ragas: Automated Evaluation of Retrieval Augmented Generation,https://arxiv.org/abs/2309.15217,2309.15217,arXiv,2023,9,arxiv_id,academic, meirtz_context,1832,Evaluation Paradigms for Context-Driven Systems > Context Quality Assessment,Human Evaluation Protocol for Generative AI Chatbots in Clinical Microbiology,https://doi.org/10.1371/journal.pone.0300487,,PLOS,2024,3,badge,academic, meirtz_context,1851,Evaluation Paradigms for Context-Driven Systems > Benchmarking Context Engineering,Agent-Pro: Learning to Evolve Coder Agents via Proposal-based Programming,https://arxiv.org/abs/2402.17574,2402.17574,arXiv,2024,2,arxiv_id,academic, meirtz_context,1853,Evaluation Paradigms for Context-Driven Systems > Benchmarking Context Engineering,GenoTEX: An LLM Agent Benchmark for Automated Gene Expression Data Analysis,https://arxiv.org/abs/2406.15341,2406.15341,MLCB,2024,6,arxiv_id,academic, meirtz_context,1869,Evaluation Paradigms for Context-Driven Systems > Agent Observability and Telemetry,LangSmith Observability Quickstart,https://docs.langchain.com/langsmith/observability-quickstart,,LangChain,,,,grey, meirtz_context,1870,Evaluation Paradigms for Context-Driven Systems > Agent Observability and Telemetry,OpenTelemetry Semantic Conventions for Generative AI,https://opentelemetry.io/docs/specs/semconv/gen-ai/,,OpenTelemetry,,,,grey, meirtz_context,1871,Evaluation Paradigms for Context-Driven Systems > Agent Observability and Telemetry,Google ADK Evaluation and Observability,https://google.github.io/adk-docs/,,Google,,,,grey, meirtz_context,1872,Evaluation Paradigms for Context-Driven Systems > Agent Observability and Telemetry,OpenAI Agents and Tools,https://platform.openai.com/docs/guides/agents,,OpenAI,,,,grey, meirtz_context,1884,Applications and Systems > Complex Research Systems,Hypothesis Generation with Large Language Models,https://arxiv.org/abs/2404.04326,2404.04326,arXiv,2024,4,arxiv_id,academic, meirtz_context,1888,Applications and Systems > Complex Research Systems,GFlowNets for AI-Driven Scientific Discovery,https://doi.org/10.1039/D3DD00002H,,Digital_Discovery,2023,6,badge,academic, meirtz_context,1890,Applications and Systems > Complex Research Systems,Literature Meets Data: A Synergistic Approach to Hypothesis Generation,https://arxiv.org/abs/2410.17309,2410.17309,arXiv,2024,10,arxiv_id,academic, meirtz_context,1894,Applications and Systems > Complex Research Systems,Machine Learning for Hypothesis Generation in Biology and Medicine,https://doi.org/10.1039/D3DD00185G,,Digital_Discovery,2024,2,badge,academic, meirtz_context,1901,Applications and Systems > Complex Research Systems,The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery,https://arxiv.org/abs/2408.06292,2408.06292,arXiv,2024,8,arxiv_id,academic, meirtz_context,1905,Applications and Systems > Complex Research Systems,Automating Psychological Hypothesis Generation with AI,https://doi.org/10.1038/s41599-024-03407-5,,Nature,2024,7,badge,academic, meirtz_context,1907,Applications and Systems > Complex Research Systems,Can Large Language Models Replace Humans in Systematic Reviews?,https://doi.org/10.1002/jrsm.1715,,Research_Synthesis,,,,academic, meirtz_context,1909,Applications and Systems > Complex Research Systems,Solving Olympiad Geometry without Human Demonstrations,https://www.nature.com/articles/s41586-023-06747-5,,Nature,2024,1,badge,academic, meirtz_context,1911,Applications and Systems > Complex Research Systems,GenoMAS: A Multi-Agent Framework for Scientific Discovery via Code-Driven Gene Expression Analysis,https://arxiv.org/abs/2507.21035,2507.21035,arXiv,2025,7,arxiv_id,academic, meirtz_context,1915,Applications and Systems > Complex Research Systems,aiXiv: A Next-Generation Open Access Ecosystem for Scientific Discovery Generated by AI Scientists,https://arxiv.org/abs/2508.15126,2508.15126,arXiv,2025,8,arxiv_id,academic, meirtz_context,1924,Applications and Systems > Complex Research Systems,AI for Science 2025: Convergence of AI Innovation and Scientific Discovery,https://doi.org/10.1038/d41573-025-00161-3,,Nature,2025,5,badge,academic, meirtz_context,1925,Applications and Systems > Complex Research Systems,"Towards Scientific Discovery with Generative AI: Progress, Opportunities, and Challenges",https://arxiv.org/abs/2412.11427,2412.11427,arXiv,2024,12,arxiv_id,academic, meirtz_context,1930,Applications and Systems > Complex Research Systems,Accelerating scientific discovery with AI,https://news.mit.edu/2025/futurehouse-accelerates-scientific-discovery-with-ai-0630,,MIT,2025,6,badge,grey, meirtz_context,1932,Applications and Systems > Complex Research Systems,Accelerating scientific breakthroughs with an AI co-scientist,https://research.google/blog/accelerating-scientific-breakthroughs-with-an-ai-co-scientist/,,Google,,,,grey, meirtz_context,1934,Applications and Systems > Complex Research Systems,Bridging AI and Science: Implications from a Large-Scale Literature Analysis of AI4Science,https://arxiv.org/abs/2412.09628,2412.09628,arXiv,2024,12,arxiv_id,academic, meirtz_context,1938,Applications and Systems > Complex Research Systems,AI for scientific discovery,https://www.weforum.org/publications/top-10-emerging-technologies-2024/in-full/1-ai-for-scientific-discovery/,,WEF,,,,grey, meirtz_context,1947,Applications and Systems > Production Systems,From Prompt Craft to System Design: Context Engineering as a Core Discipline for AI-Driven Delivery,https://fortegrp.com/insights/context-engineering-as-a-core-discipline-for-ai-driven-delivery,,Forte,2025,7,badge,grey, meirtz_context,1948,Applications and Systems > Production Systems,Context Engineering: A Framework for Enterprise AI Operations,https://shellypalmer.com/2025/06/context-engineering-a-framework-for-enterprise-ai-operations/,,ShellyPalmer,2025,6,badge,grey, meirtz_context,1949,Applications and Systems > Production Systems,How MCP Handles Context Management in High-Throughput Scenarios,https://portkey.ai/blog/model-context-protocol-context-management-in-high-throughput,,Portkey,2025,3,badge,grey, meirtz_context,1954,Applications and Systems > Production Systems,Case Study: JPMorgan's COiN Platform – Agentic AI for Financial Analysis,https://www.ai-mindset.ai/enterprise-ai-case-studies#JPMorgan,,Banking,2025,2,badge,grey, meirtz_context,1955,Applications and Systems > Production Systems,Case Study: EY's Agentic AI Integration in Microsoft 365 Copilot,https://www.ai-mindset.ai/enterprise-ai-case-studies#EY,,Professional_Services,2025,2,badge,grey, meirtz_context,1956,Applications and Systems > Production Systems,Context Is Everything: The Massive Shift Making AI Actually Work in the Real World,https://www.philmora.com/the-big-picture/context-is-everything-the-massive-shift-making-ai-actually-work-in-the-real-world,,Cross_Industry,2025,6,badge,grey, meirtz_context,1961,Applications and Systems > Production Systems,The Context Layer for Enterprise RAG Applications,https://contextual.ai,,Contextual_AI,2025,7,badge,grey, meirtz_context,1962,Applications and Systems > Production Systems,Navigating AI Model Deployment: Challenges and Solutions,https://www.linkedin.com/pulse/from-poc-production-overcoming-ai-deployment-ensuring-dean-lancaster-fmtoe,,LinkedIn,2025,3,badge,grey, meirtz_context,1963,Applications and Systems > Production Systems,2024: The State of Generative AI in the Enterprise,https://menlovc.com/2024-the-state-of-generative-ai-in-the-enterprise/,,Report,,,,grey, meirtz_context,1964,Applications and Systems > Production Systems,How 100 Enterprise CIOs Are Building and Buying Gen AI in 2025,https://a16z.com/ai-enterprise-2025/,,a16z,,,,grey, meirtz_context,1969,Applications and Systems > Production Systems,"Autohand Code CLI: Autonomous Coding Agent with Semantic Search, Memory, and Context Management",https://www.autohand.ai/code/,,Tool,,,,grey, meirtz_context,1980,Applications and Systems > Production Systems > Coding Agents and Project Memory,Introducing Codex,https://openai.com/index/introducing-codex/,,OpenAI,2025,5,badge,grey, meirtz_context,1981,Applications and Systems > Production Systems > Coding Agents and Project Memory,Claude Code Memory,https://docs.anthropic.com/en/docs/claude-code/memory,,Anthropic,,,,grey, meirtz_context,1982,Applications and Systems > Production Systems > Coding Agents and Project Memory,Claude Code Subagents,https://docs.anthropic.com/en/docs/claude-code/sub-agents,,Anthropic,,,,grey, meirtz_context,1983,Applications and Systems > Production Systems > Coding Agents and Project Memory,Letta Memory Blocks,https://docs.letta.com/guides/core-concepts/memory/memory-blocks,,Letta,,,,grey, meirtz_context,1984,Applications and Systems > Production Systems > Coding Agents and Project Memory,LangChain Deep Agents,https://docs.langchain.com/oss/python/deepagents/overview,,LangChain,,,,grey, meirtz_context,1985,Applications and Systems > Production Systems > Coding Agents and Project Memory,nv:context,https://skills.nichevlabs.com,,Tool,,,,grey, meirtz_context,1993,Applications and Systems > Production Systems > Platform Stacks and Hosted Agent Runtimes,OpenAI Agents Guide,https://platform.openai.com/docs/guides/agents,,OpenAI,,,,grey, meirtz_context,1994,Applications and Systems > Production Systems > Platform Stacks and Hosted Agent Runtimes,Google Agent Development Kit (ADK),https://google.github.io/adk-docs/,,Google,,,,grey, meirtz_context,1995,Applications and Systems > Production Systems > Platform Stacks and Hosted Agent Runtimes,Vertex AI Agent Engine,https://cloud.google.com/vertex-ai/generative-ai/docs/agent-engine/overview,,Google,,,,grey, meirtz_context,1996,Applications and Systems > Production Systems > Platform Stacks and Hosted Agent Runtimes,LangGraph Memory Overview,https://docs.langchain.com/oss/javascript/langgraph/memory,,LangChain,,,,grey, meirtz_context,1997,Applications and Systems > Production Systems > Platform Stacks and Hosted Agent Runtimes,Microsoft Agent Framework,https://learn.microsoft.com/en-us/agent-framework/user-guide/overview,,Microsoft,,,,grey, meirtz_context,1998,Applications and Systems > Production Systems > Platform Stacks and Hosted Agent Runtimes,Puppyone,https://github.com/puppyone-ai/puppyone,,GitHub,,,,grey, deepjlu_graph,95,Related Survey Papers,Graph Retrieval-Augmented Generation: A Survey,https://arxiv.org/abs/2408.08921,2408.08921,arXiv,2024,8,arxiv_id,academic, deepjlu_graph,96,Related Survey Papers,Agent Harness Engineering: A Survey,https://openreview.net/forum?id=eONq7FdiHa,,arXiv,2026,,venue_tag,academic, deepjlu_graph,97,Related Survey Papers,A Systematic Survey of Self-Evolving Agents: From Model-Centric to Environment-Driven Co-Evolution,https://www.techrxiv.org/doi/full/10.36227/techrxiv.177203250.05832634/v2,,TechRxiv,2026,,venue_tag,academic, deepjlu_graph,98,Related Survey Papers,"Graphs Meet AI Agents: Taxonomy, Progress, and Future Opportunities",https://arxiv.org/abs/2506.18019,2506.18019,arXiv,2025,6,arxiv_id,academic, deepjlu_graph,99,Related Survey Papers,Graph-Augmented Large Language Model Agents: Current Progress and Future Prospects,https://arxiv.org/abs/2507.21407,2507.21407,arXiv,2025,7,arxiv_id,academic, deepjlu_graph,100,Related Survey Papers,"Integrating Graphs, Large Language Models, and Agents: Reasoning and Retrieval",https://arxiv.org/abs/2604.15951,2604.15951,arXiv,2026,4,arxiv_id,academic, deepjlu_graph,101,Related Survey Papers,Understanding the Planning of LLM Agents: A Survey,https://arxiv.org/abs/2402.02716,2402.02716,arXiv,2024,2,arxiv_id,academic, deepjlu_graph,102,Related Survey Papers,A Survey on Agent Workflow—Status and Future,https://arxiv.org/abs/2508.01186,2508.01186,Paper,2025,8,arxiv_id,academic, deepjlu_graph,103,Related Survey Papers,From Agent Loops to Structured Graphs: A Scheduler-Theoretic Framework for LLM Agent Execution,https://arxiv.org/abs/2604.11378,2604.11378,arXiv,2026,4,arxiv_id,academic, deepjlu_graph,104,Related Survey Papers,"A Survey on LLM-Based Multi-Agent Systems: Workflow, Infrastructure, and Challenges",https://link.springer.com/article/10.1007/s44336-024-00009-2,,Vicinagearth,2024,,venue_tag,academic, deepjlu_graph,105,Related Survey Papers,Beyond Self-Talk: A Communication-Centric Survey of LLM-Based Multi-Agent Systems,https://arxiv.org/abs/2502.14321,2502.14321,Frontiers of Computer Science,2025,2,arxiv_id,academic, deepjlu_graph,106,Related Survey Papers,"Graph-Based Agent Memory: Taxonomy, Techniques, and Applications",https://arxiv.org/abs/2602.05665,2602.05665,arXiv,2026,2,arxiv_id,academic, deepjlu_graph,107,Related Survey Papers,"A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence",https://arxiv.org/abs/2507.21046,2507.21046,TMLR,2025,7,arxiv_id,academic, deepjlu_graph,108,Related Survey Papers,Self-Improving Agents in the Era of Experience: A Survey of Self- to Meta-Evolution,https://openreview.net/forum?id=IUltZSgLMm,,OpenReview Archive,2026,,venue_tag,academic, deepjlu_graph,109,Related Survey Papers,Multi-Agent Collaboration Mechanisms: A Survey of LLMs,https://arxiv.org/abs/2501.06322,2501.06322,arXiv,2025,1,arxiv_id,academic, deepjlu_graph,110,Related Survey Papers,"Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems",https://arxiv.org/abs/2605.14892,2605.14892,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,111,Related Survey Papers,"The Agent Loop: A Survey of Control Strategies, Skills, and Harnesses for LLM Agents",https://ssrn.com/abstract=7186738,,SSRN,2026,,venue_tag,academic, deepjlu_graph,130,Research Papers > Model Intelligence > Pre-Training,Language Models are Few-Shot Learners,https://scholar.google.com/scholar?q=Language+Models+are+Few-Shot+Learners,,NeurIPS,2020,,venue_tag,academic, deepjlu_graph,131,Research Papers > Model Intelligence > Pre-Training,"Scaling Language Models: Methods, Analysis & Insights from Training Gopher",https://arxiv.org/abs/2112.11446,2112.11446,arXiv,2021,12,arxiv_id,academic, deepjlu_graph,132,Research Papers > Model Intelligence > Pre-Training,PaLM: Scaling Language Modeling with Pathways,https://scholar.google.com/scholar?q=PaLM%3A+Scaling+Language+Modeling+with+Pathways,,JMLR,2023,,venue_tag,academic, deepjlu_graph,133,Research Papers > Model Intelligence > Pre-Training,LLaMA: Open and Efficient Foundation Language Models,https://arxiv.org/abs/2302.13971,2302.13971,arXiv,2023,2,arxiv_id,academic, deepjlu_graph,134,Research Papers > Model Intelligence > Pre-Training,Scaling Laws for Neural Language Models,https://arxiv.org/abs/2001.08361,2001.08361,arXiv,2020,1,arxiv_id,academic, deepjlu_graph,135,Research Papers > Model Intelligence > Pre-Training,Training Compute-Optimal Large Language Models,https://scholar.google.com/scholar?q=Training+Compute-Optimal+Large+Language+Models,,NeurIPS,2022,,venue_tag,academic, deepjlu_graph,136,Research Papers > Model Intelligence > Pre-Training,Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity,https://scholar.google.com/scholar?q=Switch+Transformers%3A+Scaling+to+Trillion+Parameter+Models+with+Simple+and+Efficient+Sparsity,,JMLR,2022,,venue_tag,academic, deepjlu_graph,137,Research Papers > Model Intelligence > Pre-Training,Mixtral of Experts,https://arxiv.org/abs/2401.04088,2401.04088,arXiv,2024,1,arxiv_id,academic, deepjlu_graph,138,Research Papers > Model Intelligence > Pre-Training,DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models,https://scholar.google.com/scholar?q=DeepSeekMoE%3A+Towards+Ultimate+Expert+Specialization+in+Mixture-of-Experts+Language+Models,,Paper,2024,,venue_tag,academic, deepjlu_graph,139,Research Papers > Model Intelligence > Pre-Training,The Llama 3 Herd of Models,https://arxiv.org/abs/2407.21783,2407.21783,arXiv,2024,7,arxiv_id,academic, deepjlu_graph,140,Research Papers > Model Intelligence > Pre-Training,DeepSeek-V3 Technical Report,https://arxiv.org/abs/2412.19437,2412.19437,arXiv,2024,12,arxiv_id,academic, deepjlu_graph,141,Research Papers > Model Intelligence > Pre-Training,Deduplicating Training Data Makes Language Models Better,https://scholar.google.com/scholar?q=Deduplicating+Training+Data+Makes+Language+Models+Better,,Paper,2022,,venue_tag,academic, deepjlu_graph,142,Research Papers > Model Intelligence > Pre-Training,The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale,https://scholar.google.com/scholar?q=The+FineWeb+Datasets%3A+Decanting+the+Web+for+the+Finest+Text+Data+at+Scale,,NeurIPS,2024,,venue_tag,academic, deepjlu_graph,143,Research Papers > Model Intelligence > Pre-Training,DataComp-LM: In Search of the Next Generation of Training Sets for Language Models,https://scholar.google.com/scholar?q=DataComp-LM%3A+In+Search+of+the+Next+Generation+of+Training+Sets+for+Language+Models,,NeurIPS,2024,,venue_tag,academic, deepjlu_graph,144,Research Papers > Model Intelligence > Pre-Training,Qwen2.5 Technical Report,https://arxiv.org/abs/2412.15115,2412.15115,arXiv,2024,12,arxiv_id,academic, deepjlu_graph,145,Research Papers > Model Intelligence > Pre-Training,Qwen3 Technical Report,https://arxiv.org/abs/2505.09388,2505.09388,arXiv,2025,5,arxiv_id,academic, deepjlu_graph,146,Research Papers > Model Intelligence > Pre-Training,Kimi K2: Open Agentic Intelligence,https://arxiv.org/abs/2507.20534,2507.20534,arXiv,2025,7,arxiv_id,academic, deepjlu_graph,150,Research Papers > Model Intelligence > Post-Training,Finetuned Language Models Are Zero-Shot Learners,https://scholar.google.com/scholar?q=Finetuned+Language+Models+Are+Zero-Shot+Learners,,ICLR,2022,,venue_tag,academic, deepjlu_graph,151,Research Papers > Model Intelligence > Post-Training,Multitask Prompted Training Enables Zero-Shot Task Generalization,https://scholar.google.com/scholar?q=Multitask+Prompted+Training+Enables+Zero-Shot+Task+Generalization,,ICLR,2022,,venue_tag,academic, deepjlu_graph,152,Research Papers > Model Intelligence > Post-Training,Training Language Models to Follow Instructions with Human Feedback,https://scholar.google.com/scholar?q=Training+Language+Models+to+Follow+Instructions+with+Human+Feedback,,NeurIPS,2022,,venue_tag,academic, deepjlu_graph,153,Research Papers > Model Intelligence > Post-Training,The Flan Collection: Designing Data and Methods for Effective Instruction Tuning,https://scholar.google.com/scholar?q=The+Flan+Collection%3A+Designing+Data+and+Methods+for+Effective+Instruction+Tuning,,ICML,2023,,venue_tag,academic, deepjlu_graph,154,Research Papers > Model Intelligence > Post-Training,Constitutional AI: Harmlessness from AI Feedback,https://arxiv.org/abs/2212.08073,2212.08073,arXiv,2022,12,arxiv_id,academic, deepjlu_graph,155,Research Papers > Model Intelligence > Post-Training,RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback,https://arxiv.org/abs/2309.00267,2309.00267,arXiv,2023,9,arxiv_id,academic, deepjlu_graph,156,Research Papers > Model Intelligence > Post-Training,Direct Preference Optimization: Your Language Model is Secretly a Reward Model,https://scholar.google.com/scholar?q=Direct+Preference+Optimization%3A+Your+Language+Model+is+Secretly+a+Reward+Model,,NeurIPS,2023,,venue_tag,academic, deepjlu_graph,157,Research Papers > Model Intelligence > Post-Training,Tulu 3: Pushing Frontiers in Open Language Model Post-Training,https://arxiv.org/abs/2411.15124,2411.15124,arXiv,2024,11,arxiv_id,academic, deepjlu_graph,158,Research Papers > Model Intelligence > Post-Training,DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models,https://arxiv.org/abs/2402.03300,2402.03300,arXiv,2024,2,arxiv_id,academic, deepjlu_graph,159,Research Papers > Model Intelligence > Post-Training,DeepSeek-R1 Incentivizes Reasoning in LLMs through Reinforcement Learning,https://scholar.google.com/scholar?q=DeepSeek-R1+Incentivizes+Reasoning+in+LLMs+through+Reinforcement+Learning,,Nature,2025,,venue_tag,academic, deepjlu_graph,160,Research Papers > Model Intelligence > Post-Training,DAPO: An Open-Source LLM Reinforcement Learning System at Scale,https://arxiv.org/abs/2503.14476,2503.14476,arXiv,2025,3,arxiv_id,academic, deepjlu_graph,161,Research Papers > Model Intelligence > Post-Training,WebRL: Training LLM Web Agents via Self-Evolving Online Curriculum Reinforcement Learning,https://scholar.google.com/scholar?q=WebRL%3A+Training+LLM+Web+Agents+via+Self-Evolving+Online+Curriculum+Reinforcement+Learning,,ICLR,2025,,venue_tag,academic, deepjlu_graph,162,Research Papers > Model Intelligence > Post-Training,Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning,https://arxiv.org/abs/2503.09516,2503.09516,arXiv,2025,3,arxiv_id,academic, deepjlu_graph,163,Research Papers > Model Intelligence > Post-Training,ReTool: Reinforcement Learning for Strategic Tool Use in LLMs,https://arxiv.org/abs/2504.11536,2504.11536,arXiv,2025,4,arxiv_id,academic, deepjlu_graph,164,Research Papers > Model Intelligence > Post-Training,ToolRL: Reward is All Tool Learning Needs,https://arxiv.org/abs/2504.13958,2504.13958,arXiv,2025,4,arxiv_id,academic, deepjlu_graph,165,Research Papers > Model Intelligence > Post-Training,RAGEN: Understanding Self-Evolution in LLM Agents via Multi-Turn Reinforcement Learning,https://arxiv.org/abs/2504.20073,2504.20073,arXiv,2025,4,arxiv_id,academic, deepjlu_graph,166,Research Papers > Model Intelligence > Post-Training,Agent-R1: Training Powerful LLM Agents with End-to-End Reinforcement Learning,https://arxiv.org/abs/2511.14460,2511.14460,arXiv,2025,11,arxiv_id,academic, deepjlu_graph,167,Research Papers > Model Intelligence > Post-Training,Agent Lightning: Train ANY AI Agents with Reinforcement Learning,https://arxiv.org/abs/2508.03680,2508.03680,arXiv,2025,8,arxiv_id,academic, deepjlu_graph,168,Research Papers > Model Intelligence > Post-Training,DynaWeb: Model-Based Reinforcement Learning of Web Agents,https://arxiv.org/abs/2601.22149,2601.22149,arXiv,2026,1,arxiv_id,academic, deepjlu_graph,170,Research Papers > Model Intelligence > Post-Training,Training Agents to Evolve with Their Harness: TaoLive Digital Avatar Agent Technical Report,https://arxiv.org/abs/2608.15763,2608.15763,arXiv,2026,8,arxiv_id,academic, deepjlu_graph,174,Research Papers > Model Intelligence > Prompt Engineering,Prompt Programming for Large Language Models: Beyond the Few-Shot Paradigm,https://arxiv.org/abs/2102.07350,2102.07350,arXiv,2021,2,arxiv_id,academic, deepjlu_graph,175,Research Papers > Model Intelligence > Prompt Engineering,Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?,https://scholar.google.com/scholar?q=Rethinking+the+Role+of+Demonstrations%3A+What+Makes+In-Context+Learning+Work%3F,,Paper,2022,,venue_tag,academic, deepjlu_graph,176,Research Papers > Model Intelligence > Prompt Engineering,What Makes Good In-Context Examples for GPT-3?,https://scholar.google.com/scholar?q=What+Makes+Good+In-Context+Examples+for+GPT-3%3F,,Paper,2022,,venue_tag,academic, deepjlu_graph,177,Research Papers > Model Intelligence > Prompt Engineering,Chain-of-Thought Prompting Elicits Reasoning in Large Language Models,https://scholar.google.com/scholar?q=Chain-of-Thought+Prompting+Elicits+Reasoning+in+Large+Language+Models,,NeurIPS,2022,,venue_tag,academic, deepjlu_graph,178,Research Papers > Model Intelligence > Prompt Engineering,Self-Consistency Improves Chain of Thought Reasoning in Language Models,https://scholar.google.com/scholar?q=Self-Consistency+Improves+Chain+of+Thought+Reasoning+in+Language+Models,,ICLR,2023,,venue_tag,academic, deepjlu_graph,179,Research Papers > Model Intelligence > Prompt Engineering,Least-to-Most Prompting Enables Complex Reasoning in Large Language Models,https://scholar.google.com/scholar?q=Least-to-Most+Prompting+Enables+Complex+Reasoning+in+Large+Language+Models,,ICLR,2023,,venue_tag,academic, deepjlu_graph,180,Research Papers > Model Intelligence > Prompt Engineering,Tree of Thoughts: Deliberate Problem Solving with Large Language Models,https://scholar.google.com/scholar?q=Tree+of+Thoughts%3A+Deliberate+Problem+Solving+with+Large+Language+Models,,NeurIPS,2023,,venue_tag,academic, deepjlu_graph,181,Research Papers > Model Intelligence > Prompt Engineering,Self-Refine: Iterative Refinement with Self-Feedback,https://scholar.google.com/scholar?q=Self-Refine%3A+Iterative+Refinement+with+Self-Feedback,,NeurIPS,2023,,venue_tag,academic, deepjlu_graph,182,Research Papers > Model Intelligence > Prompt Engineering,Graph of Thoughts: Solving Elaborate Problems with Large Language Models,https://scholar.google.com/scholar?q=Graph+of+Thoughts%3A+Solving+Elaborate+Problems+with+Large+Language+Models,,AAAI,2024,,venue_tag,academic, deepjlu_graph,183,Research Papers > Model Intelligence > Prompt Engineering,AutoPrompt: Eliciting Knowledge from Language Models with Automatically Generated Prompts,https://scholar.google.com/scholar?q=AutoPrompt%3A+Eliciting+Knowledge+from+Language+Models+with+Automatically+Generated+Prompts,,Paper,2020,,venue_tag,academic, deepjlu_graph,184,Research Papers > Model Intelligence > Prompt Engineering,Large Language Models Are Human-Level Prompt Engineers,https://scholar.google.com/scholar?q=Large+Language+Models+Are+Human-Level+Prompt+Engineers,,ICLR,2023,,venue_tag,academic, deepjlu_graph,185,Research Papers > Model Intelligence > Prompt Engineering,Large Language Models as Optimizers,https://scholar.google.com/scholar?q=Large+Language+Models+as+Optimizers,,ICLR,2024,,venue_tag,academic, deepjlu_graph,186,Research Papers > Model Intelligence > Prompt Engineering,Promptbreeder: Self-Referential Self-Improvement via Prompt Evolution,https://scholar.google.com/scholar?q=Promptbreeder%3A+Self-Referential+Self-Improvement+via+Prompt+Evolution,,ICLR,2024,,venue_tag,academic, deepjlu_graph,187,Research Papers > Model Intelligence > Prompt Engineering,TextGrad: Automatic Differentiation'' via Text,https://arxiv.org/abs/2406.07496,2406.07496,arXiv,2024,6,arxiv_id,academic, deepjlu_graph,191,Research Papers > Model Intelligence > Context Engineering,Dense Passage Retrieval for Open-Domain Question Answering,https://scholar.google.com/scholar?q=Dense+Passage+Retrieval+for+Open-Domain+Question+Answering,,Paper,2020,,venue_tag,academic, deepjlu_graph,192,Research Papers > Model Intelligence > Context Engineering,Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks,https://scholar.google.com/scholar?q=Retrieval-Augmented+Generation+for+Knowledge-Intensive+NLP+Tasks,,NeurIPS,2020,,venue_tag,academic, deepjlu_graph,193,Research Papers > Model Intelligence > Context Engineering,Leveraging Passage Retrieval with Generative Models for Open Domain Question Answering,https://scholar.google.com/scholar?q=Leveraging+Passage+Retrieval+with+Generative+Models+for+Open+Domain+Question+Answering,,Paper,2021,,venue_tag,academic, deepjlu_graph,194,Research Papers > Model Intelligence > Context Engineering,Precise Zero-Shot Dense Retrieval without Relevance Labels,https://scholar.google.com/scholar?q=Precise+Zero-Shot+Dense+Retrieval+without+Relevance+Labels,,Paper,2023,,venue_tag,academic, deepjlu_graph,195,Research Papers > Model Intelligence > Context Engineering,Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questions,https://scholar.google.com/scholar?q=Interleaving+Retrieval+with+Chain-of-Thought+Reasoning+for+Knowledge-Intensive+Multi-Step+Questions,,Paper,2023,,venue_tag,academic, deepjlu_graph,196,Research Papers > Model Intelligence > Context Engineering,"Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection",https://scholar.google.com/scholar?q=Self-RAG%3A+Learning+to+Retrieve%2C+Generate%2C+and+Critique+through+Self-Reflection,,ICLR,2024,,venue_tag,academic, deepjlu_graph,197,Research Papers > Model Intelligence > Context Engineering,Chain-of-Retrieval Augmented Generation,https://scholar.google.com/scholar?q=Chain-of-Retrieval+Augmented+Generation,,NeurIPS,2025,,venue_tag,academic, deepjlu_graph,198,Research Papers > Model Intelligence > Context Engineering,RankRAG: Unifying Context Ranking with Retrieval-Augmented Generation in LLMs,https://scholar.google.com/scholar?q=RankRAG%3A+Unifying+Context+Ranking+with+Retrieval-Augmented+Generation+in+LLMs,,NeurIPS,2024,,venue_tag,academic, deepjlu_graph,199,Research Papers > Model Intelligence > Context Engineering,LLMLingua: Compressing Prompts for Accelerated Inference of Large Language Models,https://scholar.google.com/scholar?q=LLMLingua%3A+Compressing+Prompts+for+Accelerated+Inference+of+Large+Language+Models,,Paper,2023,,venue_tag,academic, deepjlu_graph,200,Research Papers > Model Intelligence > Context Engineering,RECOMP: Improving Retrieval-Augmented LMs with Compression and Selective Augmentation,https://scholar.google.com/scholar?q=RECOMP%3A+Improving+Retrieval-Augmented+LMs+with+Compression+and+Selective+Augmentation,,ICLR,2024,,venue_tag,academic, deepjlu_graph,201,Research Papers > Model Intelligence > Context Engineering,From Local to Global: A Graph RAG Approach to Query-Focused Summarization,https://arxiv.org/abs/2404.16130,2404.16130,arXiv,2024,4,arxiv_id,academic, deepjlu_graph,202,Research Papers > Model Intelligence > Context Engineering,Provence: Efficient and Robust Context Pruning for Retrieval-Augmented Generation,https://scholar.google.com/scholar?q=Provence%3A+Efficient+and+Robust+Context+Pruning+for+Retrieval-Augmented+Generation,,ICLR,2025,,venue_tag,academic, deepjlu_graph,203,Research Papers > Model Intelligence > Context Engineering,Lost in the Middle: How Language Models Use Long Contexts,https://scholar.google.com/scholar?q=Lost+in+the+Middle%3A+How+Language+Models+Use+Long+Contexts,,Paper,2024,,venue_tag,academic, deepjlu_graph,204,Research Papers > Model Intelligence > Context Engineering,MemGPT: Towards LLMs as Operating Systems,https://arxiv.org/abs/2310.08560,2310.08560,arXiv,2023,10,arxiv_id,academic, deepjlu_graph,205,Research Papers > Model Intelligence > Context Engineering,HiAgent: Hierarchical Working Memory Management for Solving Long-Horizon Agent Tasks with Large Language Model,https://arxiv.org/abs/2408.09559,2408.09559,arXiv,2024,8,arxiv_id,academic, deepjlu_graph,206,Research Papers > Model Intelligence > Context Engineering,ACON: Optimizing Context Compression for Long-horizon LLM Agents,https://arxiv.org/abs/2510.00615,2510.00615,ICML,2025,10,arxiv_id,academic, deepjlu_graph,207,Research Papers > Model Intelligence > Context Engineering,Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models,https://arxiv.org/abs/2510.04618,2510.04618,ICLR,2025,10,arxiv_id,academic, deepjlu_graph,208,Research Papers > Model Intelligence > Context Engineering,Escaping the Context Bottleneck: Active Context Curation for LLM Agents via Reinforcement Learning,https://arxiv.org/abs/2604.11462,2604.11462,arXiv,2026,4,arxiv_id,academic, deepjlu_graph,209,Research Papers > Model Intelligence > Context Engineering,Learning Agent-Compatible Context Management for Long-Horizon Tasks,https://arxiv.org/abs/2605.30785,2605.30785,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,215,Research Papers > Individual Intelligence > Tool Integration,"MRKL Systems: A Modular, Neuro-Symbolic Architecture that Combines Large Language Models, External Knowledge Sources and Discrete Reasoning",https://arxiv.org/abs/2205.00445,2205.00445,arXiv,2022,5,arxiv_id,academic, deepjlu_graph,216,Research Papers > Individual Intelligence > Tool Integration,TALM: Tool Augmented Language Models,https://arxiv.org/abs/2205.12255,2205.12255,arXiv,2022,5,arxiv_id,academic, deepjlu_graph,217,Research Papers > Individual Intelligence > Tool Integration,ReAct: Synergizing Reasoning and Acting in Language Models,https://openreview.net/forum?id=WE_vluYUL-X,,ICLR,2023,,venue_tag,academic, deepjlu_graph,218,Research Papers > Individual Intelligence > Tool Integration,Toolformer: Language Models Can Teach Themselves to Use Tools,https://scholar.google.com/scholar?q=Toolformer%3A+Language+Models+Can+Teach+Themselves+to+Use+Tools,,NeurIPS,2023,,venue_tag,academic, deepjlu_graph,219,Research Papers > Individual Intelligence > Tool Integration,API-Bank: A Comprehensive Benchmark for Tool-Augmented LLMs,https://scholar.google.com/scholar?q=API-Bank%3A+A+Comprehensive+Benchmark+for+Tool-Augmented+LLMs,,Paper,2023,,venue_tag,academic, deepjlu_graph,220,Research Papers > Individual Intelligence > Tool Integration,ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs,https://scholar.google.com/scholar?q=ToolLLM%3A+Facilitating+Large+Language+Models+to+Master+16000%2B+Real-world+APIs,,ICLR,2024,,venue_tag,academic, deepjlu_graph,221,Research Papers > Individual Intelligence > Tool Integration,Gorilla: Large Language Model Connected with Massive APIs,https://scholar.google.com/scholar?q=Gorilla%3A+Large+Language+Model+Connected+with+Massive+APIs,,NeurIPS,2024,,venue_tag,academic, deepjlu_graph,222,Research Papers > Individual Intelligence > Tool Integration,Introducing the Model Context Protocol,https://scholar.google.com/scholar?q=Introducing+the+Model+Context+Protocol,,Anthropic,2024,,venue_tag,grey, deepjlu_graph,223,Research Papers > Individual Intelligence > Tool Integration,Executable Code Actions Elicit Better LLM Agents,https://arxiv.org/abs/2402.01030,2402.01030,arXiv,2024,2,arxiv_id,academic, deepjlu_graph,224,Research Papers > Individual Intelligence > Tool Integration,SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering,https://arxiv.org/abs/2405.15793,2405.15793,arXiv,2024,5,arxiv_id,academic, deepjlu_graph,225,Research Papers > Individual Intelligence > Tool Integration,OpenHands: An Open Platform for AI Software Developers as Generalist Agents,https://arxiv.org/abs/2407.16741,2407.16741,arXiv,2024,7,arxiv_id,academic, deepjlu_graph,226,Research Papers > Individual Intelligence > Tool Integration,LLM Agents Making Agent Tools,https://arxiv.org/abs/2502.11705,2502.11705,arXiv,2025,2,arxiv_id,academic, deepjlu_graph,227,Research Papers > Individual Intelligence > Tool Integration,Introducing Codex,https://scholar.google.com/scholar?q=Introducing+Codex,,OpenAI,2025,,venue_tag,grey, deepjlu_graph,228,Research Papers > Individual Intelligence > Tool Integration,Claude 3.7 Sonnet and Claude Code,https://scholar.google.com/scholar?q=Claude+3.7+Sonnet+and+Claude+Code,,Anthropic,2025,,venue_tag,grey, deepjlu_graph,229,Research Papers > Individual Intelligence > Tool Integration,Gemini CLI: Your Open-Source AI Agent,https://scholar.google.com/scholar?q=Gemini+CLI%3A+Your+Open-Source+AI+Agent,,Google,2025,,venue_tag,grey, deepjlu_graph,230,Research Papers > Individual Intelligence > Tool Integration,GitHub Copilot: Meet the New Coding Agent,https://scholar.google.com/scholar?q=GitHub+Copilot%3A+Meet+the+New+Coding+Agent,,GitHub Blog,2025,,venue_tag,grey, deepjlu_graph,231,Research Papers > Individual Intelligence > Tool Integration,An Open-Source Spec for Codex Orchestration: Symphony,https://scholar.google.com/scholar?q=An+Open-Source+Spec+for+Codex+Orchestration%3A+Symphony,,OpenAI Engineering,2026,,venue_tag,grey, deepjlu_graph,235,Research Papers > Individual Intelligence > Memory Management,Generative Agents: Interactive Simulacra of Human Behavior,https://scholar.google.com/scholar?q=Generative+Agents%3A+Interactive+Simulacra+of+Human+Behavior,,Paper,2023,,venue_tag,academic, deepjlu_graph,236,Research Papers > Individual Intelligence > Memory Management,MemoryBank: Enhancing Large Language Models with Long-Term Memory,https://scholar.google.com/scholar?q=MemoryBank%3A+Enhancing+Large+Language+Models+with+Long-Term+Memory,,AAAI,2024,,venue_tag,academic, deepjlu_graph,237,Research Papers > Individual Intelligence > Memory Management,MemGPT: Towards LLMs as Operating Systems,https://arxiv.org/abs/2310.08560,2310.08560,arXiv,2023,10,arxiv_id,academic, deepjlu_graph,238,Research Papers > Individual Intelligence > Memory Management,A-MEM: Agentic Memory for LLM Agents,https://arxiv.org/abs/2502.12110,2502.12110,arXiv,2025,2,arxiv_id,academic, deepjlu_graph,239,Research Papers > Individual Intelligence > Memory Management,Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory,https://arxiv.org/abs/2504.19413,2504.19413,arXiv,2025,4,arxiv_id,academic, deepjlu_graph,240,Research Papers > Individual Intelligence > Memory Management,Zep: A Temporal Knowledge Graph Architecture for Agent Memory,https://arxiv.org/abs/2501.13956,2501.13956,arXiv,2025,1,arxiv_id,academic, deepjlu_graph,241,Research Papers > Individual Intelligence > Memory Management,Memory OS of AI Agent,https://arxiv.org/abs/2506.06326,2506.06326,arXiv,2025,6,arxiv_id,academic, deepjlu_graph,242,Research Papers > Individual Intelligence > Memory Management,Memoria: A Scalable Agentic Memory Framework for Personalized Conversational AI,https://arxiv.org/abs/2512.12686,2512.12686,arXiv,2025,12,arxiv_id,academic, deepjlu_graph,243,Research Papers > Individual Intelligence > Memory Management,Agentic Memory: Learning Unified Long-Term and Short-Term Memory Management for Large Language Model Agents,https://doi.org/10.18653/v1/2026.acl-long.981,,Paper,2026,,venue_tag,academic, deepjlu_graph,244,Research Papers > Individual Intelligence > Memory Management,"Memori: A Persistent Memory Layer for Efficient, Context-Aware LLM Agents",https://arxiv.org/abs/2603.19935,2603.19935,arXiv,2026,3,arxiv_id,academic, deepjlu_graph,245,Research Papers > Individual Intelligence > Memory Management,LycheeMemory V2: Efficient Long-Term Memory for LLM Agents via Semantic Segment-Level Consolidation,https://arxiv.org/abs/2608.12990,2608.12990,arXiv,2026,8,arxiv_id,academic, deepjlu_graph,246,Research Papers > Individual Intelligence > Memory Management,Agent Workflow Memory,https://arxiv.org/abs/2409.07429,2409.07429,arXiv,2024,9,arxiv_id,academic, deepjlu_graph,247,Research Papers > Individual Intelligence > Memory Management,The Context Oscillator: Why AI Memory Should Breathe,https://zenodo.org/records/21811408,,Zenodo,2026,,venue_tag,grey, deepjlu_graph,251,Research Papers > Individual Intelligence > Skill Composition,Voyager: An Open-Ended Embodied Agent with Large Language Models,https://arxiv.org/abs/2305.16291,2305.16291,arXiv,2023,5,arxiv_id,academic, deepjlu_graph,252,Research Papers > Individual Intelligence > Skill Composition,CRAFT: Customizing LLMs by Creating and Retrieving from Specialized Toolsets,https://scholar.google.com/scholar?q=CRAFT%3A+Customizing+LLMs+by+Creating+and+Retrieving+from+Specialized+Toolsets,,ICLR,2024,,venue_tag,academic, deepjlu_graph,253,Research Papers > Individual Intelligence > Skill Composition,Equipping Agents for the Real World with Agent Skills,https://scholar.google.com/scholar?q=Equipping+Agents+for+the+Real+World+with+Agent+Skills,,Anthropic Engineering,2025,,venue_tag,grey, deepjlu_graph,254,Research Papers > Individual Intelligence > Skill Composition,Reinforcement Learning for Self-Improving Agent with Skill Library,https://arxiv.org/abs/2512.17102,2512.17102,arXiv,2025,12,arxiv_id,academic, deepjlu_graph,255,Research Papers > Individual Intelligence > Skill Composition,Harnessing LLM Agents with Skill Programs,https://arxiv.org/abs/2605.17734,2605.17734,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,256,Research Papers > Individual Intelligence > Skill Composition,From Skill Text to Skill Structure: The Scheduling-Structural-Logical Representation for Agent Skills,https://arxiv.org/abs/2604.24026,2604.24026,arXiv,2026,4,arxiv_id,academic, deepjlu_graph,257,Research Papers > Individual Intelligence > Skill Composition,SkillComposer: Learning to Evolve Agent Skills for Specification and Generalization,https://arxiv.org/abs/2606.06079,2606.06079,arXiv,2026,6,arxiv_id,academic, deepjlu_graph,258,Research Papers > Individual Intelligence > Skill Composition,Generative Skill Composition for LLM Agents,https://arxiv.org/abs/2606.32025,2606.32025,arXiv,2026,6,arxiv_id,academic, deepjlu_graph,259,Research Papers > Individual Intelligence > Skill Composition,Skill-Use: Can LLMs Actually Use Skills in Agentic Harnesses?,https://arxiv.org/abs/2608.04828,2608.04828,arXiv,2026,8,arxiv_id,academic, deepjlu_graph,260,Research Papers > Individual Intelligence > Skill Composition,Hypothesis-Driven Skill Optimization for LLM Agents,https://arxiv.org/abs/2606.22330,2606.22330,arXiv,2026,6,arxiv_id,academic, deepjlu_graph,261,Research Papers > Individual Intelligence > Skill Composition,Demystifying Agent Skills: Why They Work—Until They Don't,https://arxiv.org/abs/2608.14036,2608.14036,arXiv,2026,8,arxiv_id,academic, deepjlu_graph,262,Research Papers > Individual Intelligence > Skill Composition,The Context Oscillator: Why AI Memory Should Breathe,https://zenodo.org/records/21811408,,Zenodo,2026,,venue_tag,grey, deepjlu_graph,266,Research Papers > Individual Intelligence > Runtime Orchestration,Effective Harnesses for Long-Running Agents,https://scholar.google.com/scholar?q=Effective+Harnesses+for+Long-Running+Agents,,Anthropic Engineering,2025,,venue_tag,grey, deepjlu_graph,267,Research Papers > Individual Intelligence > Runtime Orchestration,"Externalization in LLM Agents: A Unified Review of Memory, Skills, Protocols and Harness Engineering",https://arxiv.org/abs/2604.08224,2604.08224,arXiv,2026,4,arxiv_id,academic, deepjlu_graph,268,Research Papers > Individual Intelligence > Runtime Orchestration,AI Harness Engineering: A Runtime Substrate for Foundation-Model Software Agents,https://arxiv.org/abs/2605.13357,2605.13357,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,269,Research Papers > Individual Intelligence > Runtime Orchestration,Code as Agent Harness,https://arxiv.org/abs/2605.18747,2605.18747,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,270,Research Papers > Individual Intelligence > Runtime Orchestration,Configuring Agentic AI Coding Tools: An Exploratory Study,https://arxiv.org/abs/2602.14690,2602.14690,arXiv,2026,2,arxiv_id,academic, deepjlu_graph,271,Research Papers > Individual Intelligence > Runtime Orchestration,Harness-Bench: Measuring Harness Effects across Models in Realistic Agent Workflows,https://arxiv.org/abs/2605.27922,2605.27922,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,272,Research Papers > Individual Intelligence > Runtime Orchestration,From Prompts to Contracts: Harness Engineering for Auditable Enterprise LLM Agents,https://arxiv.org/abs/2607.08028,2607.08028,arXiv,2026,7,arxiv_id,academic, deepjlu_graph,273,Research Papers > Individual Intelligence > Runtime Orchestration,"ToolSandbox: A Stateful, Conversational, Interactive Evaluation Benchmark for LLM Tool Use Capabilities",https://arxiv.org/abs/2408.04682,2408.04682,arXiv,2024,8,arxiv_id,academic, deepjlu_graph,274,Research Papers > Individual Intelligence > Runtime Orchestration,Identifying the Risks of LM Agents with an LM-Emulated Sandbox,https://arxiv.org/abs/2309.15817,2309.15817,arXiv,2023,9,arxiv_id,academic, deepjlu_graph,275,Research Papers > Individual Intelligence > Runtime Orchestration,Evolving Agents in the Dark: Retrospective Harness Optimization via Self-Preference,https://arxiv.org/abs/2606.05922,2606.05922,arXiv,2026,6,arxiv_id,academic, deepjlu_graph,276,Research Papers > Individual Intelligence > Runtime Orchestration,Defeating Prompt Injections by Design,https://arxiv.org/abs/2503.18813,2503.18813,arXiv,2025,3,arxiv_id,academic, deepjlu_graph,277,Research Papers > Individual Intelligence > Runtime Orchestration,MCP Security Bench (MSB): Benchmarking Attacks Against Model Context Protocol in LLM Agents,https://arxiv.org/abs/2510.15994,2510.15994,arXiv,2025,10,arxiv_id,academic, deepjlu_graph,278,Research Papers > Individual Intelligence > Runtime Orchestration,Harness Engineering: Leveraging Codex in an Agent-First World,https://scholar.google.com/scholar?q=Harness+Engineering%3A+Leveraging+Codex+in+an+Agent-First+World,,OpenAI Engineering,2026,,venue_tag,grey, deepjlu_graph,279,Research Papers > Individual Intelligence > Runtime Orchestration,Harness Design for Long-Running Application Development,https://scholar.google.com/scholar?q=Harness+Design+for+Long-Running+Application+Development,,Anthropic Engineering,2026,,venue_tag,grey, deepjlu_graph,280,Research Papers > Individual Intelligence > Runtime Orchestration,Meta-Harness: End-to-End Optimization of Model Harnesses,https://arxiv.org/abs/2603.28052,2603.28052,arXiv,2026,3,arxiv_id,academic, deepjlu_graph,281,Research Papers > Individual Intelligence > Runtime Orchestration,Agentic Harness Engineering: Observability-Driven Automatic Evolution of Coding-Agent Harnesses,https://arxiv.org/abs/2604.25850,2604.25850,arXiv,2026,4,arxiv_id,academic, deepjlu_graph,282,Research Papers > Individual Intelligence > Runtime Orchestration,Self-Harness: Harnesses That Improve Themselves,https://arxiv.org/abs/2606.09498,2606.09498,arXiv,2026,6,arxiv_id,academic, deepjlu_graph,283,Research Papers > Individual Intelligence > Runtime Orchestration,From Failed Trajectories to Reliable LLM Agents: Diagnosing and Repairing Harness Flaws,https://arxiv.org/abs/2606.06324,2606.06324,arXiv,2026,6,arxiv_id,academic, deepjlu_graph,284,Research Papers > Individual Intelligence > Runtime Orchestration,HARBOR: Automated Harness Optimization,https://arxiv.org/abs/2604.20938,2604.20938,arXiv,2026,4,arxiv_id,academic, deepjlu_graph,285,Research Papers > Individual Intelligence > Runtime Orchestration,AgentDojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents,https://arxiv.org/abs/2406.13352,2406.13352,arXiv,2024,6,arxiv_id,academic, deepjlu_graph,286,Research Papers > Individual Intelligence > Runtime Orchestration,Harness Updating Is Not Harness Benefit: Disentangling Evolution Capabilities in Self-Evolving LLM Agents,https://arxiv.org/abs/2605.30621,2605.30621,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,287,Research Papers > Individual Intelligence > Runtime Orchestration,Adaptive Auto-Harness: Sustained Self-Improvement for Agentic System Deployment on Open-Ended Task Streams,https://arxiv.org/abs/2606.01770,2606.01770,arXiv,2026,6,arxiv_id,academic, deepjlu_graph,288,Research Papers > Individual Intelligence > Runtime Orchestration,LongHorizon-Harness: Advancing Long-Horizon Agents for Real-World Tasks,https://arxiv.org/abs/2608.01964,2608.01964,arXiv,2026,8,arxiv_id,academic, deepjlu_graph,289,Research Papers > Individual Intelligence > Runtime Orchestration,OneDayAgent: Towards a Long-Horizon Harness for Autonomous Agents,https://arxiv.org/abs/2608.05013,2608.05013,arXiv,2026,8,arxiv_id,academic, deepjlu_graph,290,Research Papers > Individual Intelligence > Runtime Orchestration,Evo-Harness: Context-to-Harness Skill Compilation for Self-Evolving Agents,https://arxiv.org/abs/2608.15071,2608.15071,arXiv,2026,8,arxiv_id,academic, deepjlu_graph,291,Research Papers > Individual Intelligence > Runtime Orchestration,"Harness Handbook: Making Evolving Agent Harnesses Readable, Navigable, and Editable",https://arxiv.org/abs/2607.13285,2607.13285,arXiv,2026,7,arxiv_id,academic, deepjlu_graph,292,Research Papers > Individual Intelligence > Runtime Orchestration,HarnessOpt-Bench: Evaluating LLMs at Harness Optimization,https://arxiv.org/abs/2608.06301,2608.06301,arXiv,2026,8,arxiv_id,academic, deepjlu_graph,293,Research Papers > Individual Intelligence > Runtime Orchestration,The Scaffold Effect in Coding Agents: Harness Choice as a Hidden Variable in Coding-Agent Evaluation,https://arxiv.org/abs/2607.22585,2607.22585,arXiv,2026,7,arxiv_id,academic, deepjlu_graph,294,Research Papers > Individual Intelligence > Runtime Orchestration,Harness-IF: Evaluating Instruction Following Across Instruction Surfaces in Coding Agents,https://arxiv.org/abs/2608.11727,2608.11727,arXiv,2026,8,arxiv_id,academic, deepjlu_graph,295,Research Papers > Individual Intelligence > Runtime Orchestration,Evo-Bench: Can Language Models Improve Agent Harness?,https://arxiv.org/abs/2608.09096,2608.09096,arXiv,2026,8,arxiv_id,academic, deepjlu_graph,299,Research Papers > Individual Intelligence > Loop Architecture,StateFlow: Enhancing LLM Task-Solving through State-Driven Workflows,https://arxiv.org/abs/2403.11322,2403.11322,arXiv,2024,3,arxiv_id,academic, deepjlu_graph,300,Research Papers > Individual Intelligence > Loop Architecture,Magentic-One: A generalist multi-agent system for solving complex tasks,https://arxiv.org/abs/2411.04468,2411.04468,arXiv,2024,11,arxiv_id,academic, deepjlu_graph,301,Research Papers > Individual Intelligence > Loop Architecture,AIOS: LLM Agent Operating System,https://arxiv.org/abs/2403.16971,2403.16971,arXiv,2024,3,arxiv_id,academic, deepjlu_graph,302,Research Papers > Individual Intelligence > Loop Architecture,AgentBoard: An Analytical Evaluation Board of Multi-turn LLM Agents,https://arxiv.org/abs/2401.13178,2401.13178,arXiv,2024,1,arxiv_id,academic, deepjlu_graph,303,Research Papers > Individual Intelligence > Loop Architecture,AdaPlanner: Adaptive Planning from Feedback with Language Models,https://arxiv.org/abs/2305.16653,2305.16653,arXiv,2023,5,arxiv_id,academic, deepjlu_graph,304,Research Papers > Individual Intelligence > Loop Architecture,When Agents Do Not Stop: Uncovering Infinite Agentic Loops in LLM Agents,https://arxiv.org/abs/2607.01641,2607.01641,arXiv,2026,7,arxiv_id,academic, deepjlu_graph,305,Research Papers > Individual Intelligence > Loop Architecture,Stop Hand-Holding Your Coding Agent: Engineering the Loops That Replace Step-by-Step Prompting,https://arxiv.org/abs/2607.00038,2607.00038,arXiv,2026,7,arxiv_id,academic, deepjlu_graph,306,Research Papers > Individual Intelligence > Loop Architecture,ResearchLoop: An Evidence-Gated Control Plane for AI-Assisted Research,https://arxiv.org/abs/2605.28282,2605.28282,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,307,Research Papers > Individual Intelligence > Loop Architecture,"Proof-or-Stop: Don't Trust the Agent, Trust the Evidence – Loop Engineering for Verifiable Evidence-Gated Lifecycle Control",https://arxiv.org/abs/2607.14890,2607.14890,arXiv,2026,7,arxiv_id,academic, deepjlu_graph,312,Research Papers > Individual Intelligence > Interaction Paradigm,Beyond Message Passing: A Semantic View of Agent Communication Protocols,https://arxiv.org/abs/2604.02369,2604.02369,arXiv,2026,4,arxiv_id,academic, deepjlu_graph,313,Research Papers > Individual Intelligence > Interaction Paradigm,"Internet of Agents: Fundamentals, Applications, and Challenges",https://arxiv.org/abs/2505.07176,2505.07176,Paper,2025,5,arxiv_id,academic, deepjlu_graph,314,Research Papers > Individual Intelligence > Interaction Paradigm,LLM Agent Communication Protocol (LACP) Requires Urgent Standardization: A Telecom-Inspired Protocol Is Necessary,https://arxiv.org/abs/2510.13821,2510.13821,NeurIPS,2025,10,arxiv_id,academic, deepjlu_graph,315,Research Papers > Individual Intelligence > Interaction Paradigm,Beyond the Protocol: Unveiling Attack Vectors in the Model Context Protocol (MCP) Ecosystem,https://doi.org/10.1109/TSE.2026.3694876,,Paper,2026,,venue_tag,academic, deepjlu_graph,316,Research Papers > Individual Intelligence > Interaction Paradigm,AgentRx: Diagnosing AI Agent Failures from Execution Trajectories,https://arxiv.org/abs/2602.02475,2602.02475,arXiv,2026,2,arxiv_id,academic, deepjlu_graph,317,Research Papers > Individual Intelligence > Interaction Paradigm,Supervising Ralph Wiggum: Exploring a Metacognitive Co-Regulation Agentic AI Loop for Engineering Design,https://arxiv.org/abs/2603.24768,2603.24768,arXiv,2026,3,arxiv_id,academic, deepjlu_graph,318,Research Papers > Individual Intelligence > Interaction Paradigm,"From Function Calls to MCPs for Securing AI Agent Systems: Architecture, Challenges and Countermeasures",https://doi.org/10.12142/ZTECOM.202503004,,ZTE Communications,2025,,venue_tag,academic, deepjlu_graph,319,Research Papers > Individual Intelligence > Interaction Paradigm,Sovereign Agentic Loops: Decoupling AI Reasoning from Execution in Real-World Systems,https://arxiv.org/abs/2604.22136,2604.22136,arXiv,2026,4,arxiv_id,academic, deepjlu_graph,320,Research Papers > Individual Intelligence > Interaction Paradigm,"The Log is the Agent: Event-Sourced Reactive Graphs for Auditable, Forkable Agentic Systems",https://arxiv.org/abs/2605.21997,2605.21997,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,324,Research Papers > Individual Intelligence > Environment Feedback,EurekAgent: Agent Environment Engineering is All You Need for Autonomous Scientific Discovery,https://arxiv.org/abs/2606.13662,2606.13662,arXiv,2026,6,arxiv_id,academic, deepjlu_graph,325,Research Papers > Individual Intelligence > Environment Feedback,LEVER: Learning to Verify Language-to-Code Generation with Execution,https://arxiv.org/abs/2302.08468,2302.08468,arXiv,2023,2,arxiv_id,academic, deepjlu_graph,326,Research Papers > Individual Intelligence > Environment Feedback,CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquing,https://arxiv.org/abs/2305.11738,2305.11738,arXiv,2023,5,arxiv_id,academic, deepjlu_graph,327,Research Papers > Individual Intelligence > Environment Feedback,Executable Code Actions Elicit Better LLM Agents,https://arxiv.org/abs/2402.01030,2402.01030,arXiv,2024,2,arxiv_id,academic, deepjlu_graph,328,Research Papers > Individual Intelligence > Environment Feedback,"ToolSandbox: A Stateful, Conversational, Interactive Evaluation Benchmark for LLM Tool Use Capabilities",https://arxiv.org/abs/2408.04682,2408.04682,arXiv,2024,8,arxiv_id,academic, deepjlu_graph,329,Research Papers > Individual Intelligence > Environment Feedback,OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments,https://arxiv.org/abs/2404.07972,2404.07972,NeurIPS,2024,4,arxiv_id,academic, deepjlu_graph,330,Research Papers > Individual Intelligence > Environment Feedback,DeltaBox: Scaling Stateful AI Agents with Millisecond-Level Sandbox Checkpoint/Rollback,https://arxiv.org/abs/2605.22781,2605.22781,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,341,Research Papers > System Intelligence > Task Organization,HuggingGPT: Solving AI Tasks with ChatGPT and Its Friends in Hugging Face,https://arxiv.org/abs/2303.17580,2303.17580,NeurIPS,2023,3,arxiv_id,academic, deepjlu_graph,342,Research Papers > System Intelligence > Task Organization,ReWOO: Decoupling Reasoning from Observations for Efficient Augmented Language Models,https://arxiv.org/abs/2305.18323,2305.18323,arXiv,2023,5,arxiv_id,academic, deepjlu_graph,343,Research Papers > System Intelligence > Task Organization,An LLM Compiler for Parallel Function Calling,https://arxiv.org/abs/2312.04511,2312.04511,ICML,2023,12,arxiv_id,academic, deepjlu_graph,344,Research Papers > System Intelligence > Task Organization,Plan-over-Graph: Towards Parallelable LLM Agent Schedule,https://arxiv.org/abs/2502.14563,2502.14563,arXiv,2025,2,arxiv_id,academic, deepjlu_graph,345,Research Papers > System Intelligence > Task Organization,TDAG: A Multi-Agent Framework Based on Dynamic Task Decomposition and Agent Generation,https://arxiv.org/abs/2402.10178,2402.10178,Neural Networks,2024,2,arxiv_id,academic, deepjlu_graph,346,Research Papers > System Intelligence > Task Organization,Flow: Modularized Agentic Workflow Automation,https://proceedings.iclr.cc/paper_files/paper/2025/hash/ba84da6921f3040b74ee163aa7451f53-Abstract-Conference.html,,ICLR,2025,,venue_tag,academic, deepjlu_graph,347,Research Papers > System Intelligence > Task Organization,VFlow: Discovering Optimal Agentic Workflows for Verilog Generation,https://arxiv.org/abs/2504.03723,2504.03723,arXiv,2025,4,arxiv_id,academic, deepjlu_graph,348,Research Papers > System Intelligence > Task Organization,GPTSwarm: Language Agents as Optimizable Graphs,https://proceedings.mlr.press/v235/zhuge24a.html,,ICML,2024,,venue_tag,academic, deepjlu_graph,349,Research Papers > System Intelligence > Task Organization,Automated Design of Agentic Systems,https://arxiv.org/abs/2408.08435,2408.08435,ICLR,2024,8,arxiv_id,academic, deepjlu_graph,350,Research Papers > System Intelligence > Task Organization,AutoFlow: Automated Workflow Generation for Large Language Model Agents,https://arxiv.org/abs/2407.12821,2407.12821,arXiv,2024,7,arxiv_id,academic, deepjlu_graph,351,Research Papers > System Intelligence > Task Organization,AFlow: Automating Agentic Workflow Generation,https://openreview.net/forum?id=z5uVAKwmjf,,ICLR,2025,,venue_tag,academic, deepjlu_graph,352,Research Papers > System Intelligence > Task Organization,A2Flow: Automating Agentic Workflow Generation via Self-Adaptive Abstraction Operators,https://arxiv.org/abs/2511.20693,2511.20693,arXiv,2025,11,arxiv_id,academic, deepjlu_graph,353,Research Papers > System Intelligence > Task Organization,MermaidFlow: Redefining Agentic Workflow Generation via Safety-Constrained Evolutionary Programming,https://arxiv.org/abs/2505.22967,2505.22967,arXiv,2025,5,arxiv_id,academic, deepjlu_graph,354,Research Papers > System Intelligence > Task Organization,DynTaskMAS: A dynamic task graph-driven framework for asynchronous and parallel llm-based multi-agent systems,https://scholar.google.com/scholar?q=Dyntaskmas%3A+A+dynamic+task+graph-driven+framework+for+asynchronous+and+parallel+llm-based+multi-agent+systems,,Paper,2025,,venue_tag,academic, deepjlu_graph,355,Research Papers > System Intelligence > Task Organization,DyFlow: Dynamic Workflow Framework for Agentic Reasoning,https://arxiv.org/abs/2509.26062,2509.26062,NeurIPS,2025,9,arxiv_id,academic, deepjlu_graph,356,Research Papers > System Intelligence > Task Organization,EvoFlow: Evolving Diverse Agentic Workflows On The Fly,https://arxiv.org/abs/2502.07373,2502.07373,arXiv,2025,2,arxiv_id,academic, deepjlu_graph,357,Research Papers > System Intelligence > Task Organization,QualityFlow: An Agentic Workflow for Program Synthesis Controlled by LLM Quality Checks,https://arxiv.org/abs/2501.17167,2501.17167,arXiv,2025,1,arxiv_id,academic, deepjlu_graph,358,Research Papers > System Intelligence > Task Organization,FlowSteer: Prompt-Only Workflow Steering Exposes Planning-Time Vulnerabilities in Multi-Agent LLM Systems,https://arxiv.org/abs/2605.11514,2605.11514,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,359,Research Papers > System Intelligence > Task Organization,"AgenticLab: A Real-World Robot Agent Platform that Can See, Think, and Act",https://arxiv.org/abs/2602.01662v1,2602.01662,arXiv,2026,2,arxiv_id,academic, deepjlu_graph,360,Research Papers > System Intelligence > Task Organization,Towards a science of scaling agent systems,https://arxiv.org/abs/2512.08296,2512.08296,arXiv,2025,12,arxiv_id,academic, deepjlu_graph,369,Research Papers > System Intelligence > Agent Coordination,"CAMEL: Communicative Agents for ""Mind"" Exploration of Large Language Model Society",https://proceedings.neurips.cc/paper/2023/hash/a3621ee907def47c1b952ade25c67698-Abstract-Conference.html,,NeurIPS,2023,,venue_tag,academic, deepjlu_graph,370,Research Papers > System Intelligence > Agent Coordination,A dynamic LLM-powered agent network for task-oriented agent collaboration,https://arxiv.org/abs/2310.02170,2310.02170,COLM,2023,10,arxiv_id,academic, deepjlu_graph,371,Research Papers > System Intelligence > Agent Coordination,Agent-oriented planning in multi-agent systems,https://arxiv.org/abs/2410.02189,2410.02189,ICLR,2024,10,arxiv_id,academic, deepjlu_graph,372,Research Papers > System Intelligence > Agent Coordination,MasRouter: Learning to route LLMs for multi-agent systems,https://arxiv.org/abs/2502.11133,2502.11133,ACL,2025,2,arxiv_id,academic, deepjlu_graph,373,Research Papers > System Intelligence > Agent Coordination,AutoAgents: A framework for automatic agent generation,https://arxiv.org/abs/2309.17288,2309.17288,IJCAI,2023,9,arxiv_id,academic, deepjlu_graph,374,Research Papers > System Intelligence > Agent Coordination,EvoAgent: Towards automatic multi-agent generation via evolutionary algorithms,https://arxiv.org/abs/2406.14228,2406.14228,NAACL,2024,6,arxiv_id,academic, deepjlu_graph,375,Research Papers > System Intelligence > Agent Coordination,Collaborative gym: A framework for enabling and evaluating human-agent collaboration,https://arxiv.org/abs/2412.15701,2412.15701,ICLR,2024,12,arxiv_id,academic, deepjlu_graph,376,Research Papers > System Intelligence > Agent Coordination,AOrchestra: Automating sub-agent creation for agentic orchestration,https://arxiv.org/abs/2602.03786,2602.03786,ICML,2026,2,arxiv_id,academic, deepjlu_graph,377,Research Papers > System Intelligence > Agent Coordination,Adaptive graph pruning for multi-agent communication,https://arxiv.org/abs/2506.02951,2506.02951,ECAI,2025,6,arxiv_id,academic, deepjlu_graph,378,Research Papers > System Intelligence > Agent Coordination,Multi-agent architecture search via agentic supernet,https://arxiv.org/abs/2502.04180,2502.04180,ICML,2025,2,arxiv_id,academic, deepjlu_graph,379,Research Papers > System Intelligence > Agent Coordination,SkillGraph: Self-Evolving Multi-Agent Collaboration with Multimodal Graph Topology,https://arxiv.org/abs/2604.17503,2604.17503,arXiv,2026,4,arxiv_id,academic, deepjlu_graph,380,Research Papers > System Intelligence > Agent Coordination,MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework,https://arxiv.org/abs/2308.00352,2308.00352,ICLR,2023,8,arxiv_id,academic, deepjlu_graph,381,Research Papers > System Intelligence > Agent Coordination,ChatDev: Communicative Agents for Software Development,https://aclanthology.org/2024.acl-long.810/,,ACL,2024,,venue_tag,academic, deepjlu_graph,382,Research Papers > System Intelligence > Agent Coordination,Magentic-One: A generalist multi-agent system for solving complex tasks,https://arxiv.org/abs/2411.04468,2411.04468,arXiv,2024,11,arxiv_id,academic, deepjlu_graph,383,Research Papers > System Intelligence > Agent Coordination,AgentVerse: Facilitating multi-agent collaboration and exploring emergent behaviors,https://arxiv.org/abs/2308.10848,2308.10848,ICLR,2023,8,arxiv_id,academic, deepjlu_graph,384,Research Papers > System Intelligence > Agent Coordination,Multi-agent collaboration via evolving orchestration,https://arxiv.org/abs/2505.19591,2505.19591,NeurIPS,2025,5,arxiv_id,academic, deepjlu_graph,385,Research Papers > System Intelligence > Agent Coordination,AgentNet: Decentralized evolutionary coordination for llm-based multi-agent systems,https://arxiv.org/abs/2504.00587,2504.00587,NeurIPS,2025,4,arxiv_id,academic, deepjlu_graph,386,Research Papers > System Intelligence > Agent Coordination,Scaling large language model-based multi-agent collaboration,https://arxiv.org/abs/2406.07155,2406.07155,ICLR,2024,6,arxiv_id,academic, deepjlu_graph,387,Research Papers > System Intelligence > Agent Coordination,Swarmagentic: Towards fully automated agentic system generation via swarm intelligence,https://arxiv.org/abs/2506.15672,2506.15672,EMNLP,2025,6,arxiv_id,academic, deepjlu_graph,388,Research Papers > System Intelligence > Agent Coordination,Mixture-of-agents enhances large language model capabilities,https://arxiv.org/abs/2406.04692,2406.04692,ICLR,2024,6,arxiv_id,academic, deepjlu_graph,389,Research Papers > System Intelligence > Agent Coordination,G-Designer: Architecting multi-agent communication topologies via graph neural networks,https://arxiv.org/abs/2410.11782,2410.11782,ICML,2024,10,arxiv_id,academic, deepjlu_graph,390,Research Papers > System Intelligence > Agent Coordination,AMAS: Adaptively determining communication topology for llm-based multi-agent system,https://arxiv.org/abs/2510.01617,2510.01617,EMNLP,2025,10,arxiv_id,academic, deepjlu_graph,391,Research Papers > System Intelligence > Agent Coordination,Cut the crap: An economical communication pipeline for llm-based multi-agent systems,https://arxiv.org/abs/2410.02506,2410.02506,ICLR,2024,10,arxiv_id,academic, deepjlu_graph,392,Research Papers > System Intelligence > Agent Coordination,AgentDropout: Dynamic agent elimination for token-efficient and high-performance llm-based multi-agent collaboration,https://arxiv.org/abs/2503.18891,2503.18891,ACL,2025,3,arxiv_id,academic, deepjlu_graph,393,Research Papers > System Intelligence > Agent Coordination,DyTopo: Dynamic topology routing for multi-agent reasoning via semantic matching,https://arxiv.org/abs/2602.06039,2602.06039,arXiv,2026,2,arxiv_id,academic, deepjlu_graph,394,Research Papers > System Intelligence > Agent Coordination,Adaptive theory of mind for LLM-based multi-agent coordination,https://arxiv.org/abs/2603.16264,2603.16264,AAAI,2026,3,arxiv_id,academic, deepjlu_graph,395,Research Papers > System Intelligence > Agent Coordination,Assemble your crew: Automatic multi-agent communication topology design via autoregressive graph generation,https://arxiv.org/abs/2507.18224,2507.18224,AAAI,2025,7,arxiv_id,academic, deepjlu_graph,396,Research Papers > System Intelligence > Agent Coordination,Learning to Generate and Extract: A Multi-Agent Collaboration Framework For Zero-shot Document-level Event Arguments Extraction,https://arxiv.org/abs/2603.02909,2603.02909,AAAI,2026,3,arxiv_id,academic, deepjlu_graph,397,Research Papers > System Intelligence > Agent Coordination,Agent-world: Scaling real-world environment synthesis for evolving general agent intelligence,https://arxiv.org/abs/2604.18292,2604.18292,arXiv,2026,4,arxiv_id,academic, deepjlu_graph,398,Research Papers > System Intelligence > Agent Coordination,Graph-of-agents: A graph-based framework for multi-agent LLM collaboration,https://arxiv.org/abs/2604.17148,2604.17148,ICLR,2026,4,arxiv_id,academic, deepjlu_graph,399,Research Papers > System Intelligence > Agent Coordination,Emergent coordination in multi-agent language models,https://arxiv.org/abs/2510.05174,2510.05174,ICLR,2025,10,arxiv_id,academic, deepjlu_graph,400,Research Papers > System Intelligence > Agent Coordination,AgentPO: Enhancing Multi-Agent Collaboration via Reinforcement Learning,https://openreview.net/forum?id=5L8uyzjn2l,,ICLR,2026,,venue_tag,academic, deepjlu_graph,401,Research Papers > System Intelligence > Agent Coordination,Multi-agent design: Optimizing agents with better prompts and topologies,https://arxiv.org/abs/2502.02533,2502.02533,ICLR,2025,2,arxiv_id,academic, deepjlu_graph,402,Research Papers > System Intelligence > Agent Coordination,Agent Discovery in Internet of Agents: Challenges and Solutions,https://arxiv.org/abs/2511.19113,2511.19113,IEEE Network,2025,11,arxiv_id,academic, deepjlu_graph,403,Research Papers > System Intelligence > Agent Coordination,LLM Agents Factory: Retrieval of Domain-Specific LLM Agents,https://arxiv.org/abs/2608.09934,2608.09934,SIGIR,2026,8,arxiv_id,academic, deepjlu_graph,404,Research Papers > System Intelligence > Agent Coordination,TacoMAS: Test-Time Co-Evolution of Topology and Capability in LLM-based Multi-Agent Systems,https://arxiv.org/abs/2605.09539,2605.09539,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,405,Research Papers > System Intelligence > Agent Coordination,EvolveRouter: Co-evolving routing and prompt for multi-agent question answering,https://arxiv.org/abs/2604.05149,2604.05149,arXiv,2026,4,arxiv_id,academic, deepjlu_graph,406,Research Papers > System Intelligence > Agent Coordination,MoRSE: Task-Oriented Multi-Agent System with Mixture of Role-Subtask Experts,https://arxiv.org/abs/2608.09251,2608.09251,arXiv,2026,8,arxiv_id,academic, deepjlu_graph,407,Research Papers > System Intelligence > Agent Coordination,Internet of agents: Weaving a web of heterogeneous agents for collaborative intelligence,https://arxiv.org/abs/2407.07061,2407.07061,ICLR,2024,7,arxiv_id,academic, deepjlu_graph,408,Research Papers > System Intelligence > Agent Coordination,"Toward an Organizational Science of Multi-Agent LLM Systems: Decoupling Who, How, and Which Algorithm",https://arxiv.org/abs/2607.25446,2607.25446,arXiv,2026,7,arxiv_id,academic, deepjlu_graph,409,Research Papers > System Intelligence > Agent Coordination,Multi-agent teams hold experts back,https://arxiv.org/abs/2602.01011,2602.01011,arXiv,2026,2,arxiv_id,academic, deepjlu_graph,410,Research Papers > System Intelligence > Agent Coordination,Dynamic role assignment for multi-agent debate,https://arxiv.org/abs/2601.17152,2601.17152,arXiv,2026,1,arxiv_id,academic, deepjlu_graph,411,Research Papers > System Intelligence > Agent Coordination,WorkTeam: Constructing Workflows from Natural Language with Multi-Agents,https://aclanthology.org/2025.naacl-industry.3/,,NAACL,2025,,venue_tag,academic, deepjlu_graph,412,Research Papers > System Intelligence > Agent Coordination,Evolve as a Team: Collaborative Self-Evolution for LLM-based Multi-Agent Systems,https://arxiv.org/abs/2605.29790,2605.29790,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,413,Research Papers > System Intelligence > Agent Coordination,WebSwarm: Recursive Multi-Agent Orchestration for Deep-and-Wide Web Search,https://arxiv.org/abs/2607.08662,2607.08662,arXiv,2026,7,arxiv_id,academic, deepjlu_graph,414,Research Papers > System Intelligence > Agent Coordination,Owl: Optimized workforce learning for general multi-agent assistance in real-world task automation,https://arxiv.org/abs/2505.23885,2505.23885,arXiv,2025,5,arxiv_id,academic, deepjlu_graph,415,Research Papers > System Intelligence > Agent Coordination,Agentsquare: Automatic llm agent search in modular design space,https://arxiv.org/abs/2410.06153,2410.06153,ICLR,2024,10,arxiv_id,academic, deepjlu_graph,416,Research Papers > System Intelligence > Agent Coordination,MAS-on-the-Fly: Dynamic Adaptation of LLM-based Multi-Agent Systems at Test Time,https://arxiv.org/abs/2602.13671,2602.13671,arXiv,2026,2,arxiv_id,academic, deepjlu_graph,417,Research Papers > System Intelligence > Agent Coordination,Drop the Hierarchy and Roles: How Self-Organizing LLM Agents Outperform Designed Structures,https://arxiv.org/abs/2603.28990,2603.28990,arXiv,2026,3,arxiv_id,academic, deepjlu_graph,418,Research Papers > System Intelligence > Agent Coordination,"Swarm Skills: A Portable, Self-Evolving Multi-Agent System Specification for Coordination Engineering",https://arxiv.org/abs/2605.10052,2605.10052,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,419,Research Papers > System Intelligence > Agent Coordination,Atlas: Adaptive trading with llm agents through dynamic prompt optimization and multi-agent coordination,https://arxiv.org/abs/2510.15949,2510.15949,arXiv,2025,10,arxiv_id,academic, deepjlu_graph,420,Research Papers > System Intelligence > Agent Coordination,"Magicore: Multi-agent, iterative, coarse-to-fine refinement for reasoning",https://aclanthology.org/2025.emnlp-main.1660/,,EMNLP,2025,,venue_tag,academic, deepjlu_graph,421,Research Papers > System Intelligence > Agent Coordination,Understanding the information propagation effects of communication topologies in llm-based multi-agent systems,https://aclanthology.org/2025.emnlp-main.623/,,EMNLP,2025,,venue_tag,academic, deepjlu_graph,422,Research Papers > System Intelligence > Agent Coordination,CARD: Towards Conditional Design of Multi-agent Topological Structures,https://arxiv.org/abs/2603.01089,2603.01089,arXiv,2026,3,arxiv_id,academic, deepjlu_graph,423,Research Papers > System Intelligence > Agent Coordination,QueenBee Planner: Skill-Evolving Communication Topologies for Token-Efficient LLM Multi-Agent Systems,https://arxiv.org/abs/2606.27492,2606.27492,arXiv,2026,6,arxiv_id,academic, deepjlu_graph,424,Research Papers > System Intelligence > Agent Coordination,Agentic aggregation for parallel scaling of long-horizon agentic tasks,https://arxiv.org/abs/2604.11753,2604.11753,arXiv,2026,4,arxiv_id,academic, deepjlu_graph,433,Research Papers > System Intelligence > State Management,StateFlow: Enhancing LLM Task-Solving through State-Driven Workflows,https://arxiv.org/abs/2403.11322,2403.11322,arXiv,2024,3,arxiv_id,academic, deepjlu_graph,434,Research Papers > System Intelligence > State Management,AutoGRAMS: Autonomous Graphical Agent Modeling Software,https://arxiv.org/abs/2407.10049,2407.10049,arXiv,2024,7,arxiv_id,academic, deepjlu_graph,435,Research Papers > System Intelligence > State Management,Magentic-One: A generalist multi-agent system for solving complex tasks,https://arxiv.org/abs/2411.04468,2411.04468,arXiv,2024,11,arxiv_id,academic, deepjlu_graph,436,Research Papers > System Intelligence > State Management,Graph of States: Solving Abductive Tasks with Large Language Models,https://icml.cc/virtual/2026/poster/65285,,ICML,2026,,venue_tag,grey, deepjlu_graph,437,Research Papers > System Intelligence > State Management,LangGraph: Low-Level Orchestration for Stateful Agents,https://github.com/langchain-ai/langgraph,,Paper,2026,,venue_tag,repository, deepjlu_graph,438,Research Papers > System Intelligence > State Management,Apache Burr: Stateful Application and Agent Framework,https://github.com/apache/burr,,Paper,2026,,venue_tag,repository, deepjlu_graph,439,Research Papers > System Intelligence > State Management,Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents,https://arxiv.org/abs/2508.19504,2508.19504,arXiv,2025,8,arxiv_id,academic, deepjlu_graph,440,Research Papers > System Intelligence > State Management,LlamaIndex Workflows: Event-Driven Agent Workflows,https://github.com/run-llama/workflows-py,,Paper,2026,,venue_tag,repository, deepjlu_graph,441,Research Papers > System Intelligence > State Management,Pydantic AI: Typed Agent Framework and Graph Runtime,https://github.com/pydantic/pydantic-ai,,Paper,2026,,venue_tag,repository, deepjlu_graph,442,Research Papers > System Intelligence > State Management,AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation,https://arxiv.org/abs/2308.08155,2308.08155,arXiv,2023,8,arxiv_id,academic, deepjlu_graph,443,Research Papers > System Intelligence > State Management,Sovereign Agentic Loops: Decoupling AI Reasoning from Execution in Real-World Systems,https://arxiv.org/abs/2604.22136,2604.22136,arXiv,2026,4,arxiv_id,academic, deepjlu_graph,444,Research Papers > System Intelligence > State Management,"Language Agent Tree Search Unifies Reasoning, Acting, and Planning in Language Models",https://arxiv.org/abs/2310.04406,2310.04406,arXiv,2023,10,arxiv_id,academic, deepjlu_graph,445,Research Papers > System Intelligence > State Management,PatchBoard: Schema-Grounded State Mutation for Reliable and Auditable LLM Multi-Agent Collaboration,https://arxiv.org/abs/2605.29313,2605.29313,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,446,Research Papers > System Intelligence > State Management,MemTX: Transactional Belief Commit for Stateful Agent Memory,https://arxiv.org/abs/2607.23929,2607.23929,arXiv,2026,7,arxiv_id,academic, deepjlu_graph,447,Research Papers > System Intelligence > State Management,Cordon: Semantic Transactions for Tool-Using LLM Agents,https://arxiv.org/abs/2606.17573,2606.17573,arXiv,2026,6,arxiv_id,academic, deepjlu_graph,448,Research Papers > System Intelligence > State Management,"Atomix: Timely, transactional tool use for reliable agentic workflows",https://arxiv.org/abs/2602.14849,2602.14849,arXiv,2026,2,arxiv_id,academic, deepjlu_graph,449,Research Papers > System Intelligence > State Management,"SagaLLM: Context Management, Validation, and Transaction Guarantees for Multi-Agent LLM Planning",https://arxiv.org/abs/2503.11951,2503.11951,arXiv,2025,3,arxiv_id,academic, deepjlu_graph,450,Research Papers > System Intelligence > State Management,Alas: Transactional and dynamic multi-agent llm planning,https://arxiv.org/abs/2511.03094,2511.03094,arXiv,2025,11,arxiv_id,academic, deepjlu_graph,451,Research Papers > System Intelligence > State Management,Robust Agent Compensation (RAC): Teaching AI Agents to Compensate,https://dl.acm.org/doi/full/10.1145/3786335.3813141,,CAIS,2026,,venue_tag,academic, deepjlu_graph,452,Research Papers > System Intelligence > State Management,ProPlay: Procedural World Models for Self-Evolving LLM Agents,https://arxiv.org/abs/2606.12780,2606.12780,arXiv,2026,6,arxiv_id,academic, deepjlu_graph,453,Research Papers > System Intelligence > State Management,DART: Semantic Recoverability for Structured Tool Agents,https://arxiv.org/abs/2605.23311,2605.23311,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,454,Research Papers > System Intelligence > State Management,AgentGit: A version control framework for reliable and scalable LLM-powered multi-agent systems,https://arxiv.org/abs/2511.00628,2511.00628,arXiv,2025,11,arxiv_id,academic, deepjlu_graph,455,Research Papers > System Intelligence > State Management,Shepherd: Enabling Programmable Meta-Agents via Reversible Agentic Execution Traces,https://arxiv.org/abs/2605.10913,2605.10913,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,456,Research Papers > System Intelligence > State Management,"The Log is the Agent: Event-Sourced Reactive Graphs for Auditable, Forkable Agentic Systems",https://arxiv.org/abs/2605.21997,2605.21997,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,457,Research Papers > System Intelligence > State Management,Verified Detection and Prevention of Concurrency Anomalies in Multi-Agent Large Language Model Systems,https://arxiv.org/abs/2606.17182,2606.17182,arXiv,2026,6,arxiv_id,academic, deepjlu_graph,458,Research Papers > System Intelligence > State Management,CausalFlow: Causal Attribution and Counterfactual Repair for LLM Agent Failures,https://arxiv.org/abs/2605.25338,2605.25338,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,459,Research Papers > System Intelligence > State Management,ReflexGrad: Within-Episode Failure Recovery in LLM Agents via Progress-Gated Dual-Process Routing,https://arxiv.org/abs/2511.14584,2511.14584,arXiv,2025,11,arxiv_id,academic, deepjlu_graph,460,Research Papers > System Intelligence > State Management,TDAD: Test-Driven Agentic Development - Reducing Code Regressions in AI Coding Agents via Graph-Based Impact Analysis,https://arxiv.org/abs/2603.17973,2603.17973,arXiv,2026,3,arxiv_id,academic, deepjlu_graph,461,Research Papers > System Intelligence > State Management,Which Agent Causes Task Failures and When? On Automated Failure Attribution of LLM Multi-Agent Systems,https://arxiv.org/abs/2505.00212,2505.00212,arXiv,2025,5,arxiv_id,academic, deepjlu_graph,462,Research Papers > System Intelligence > State Management,Why Do Multi-Agent LLM Systems Fail?,https://proceedings.neurips.cc/paper_files/paper/2025/hash/b1041e52d3be19f0a9bc491657488e4a-Abstract-Datasets_and_Benchmarks_Track.html,,NeurIPS,2025,,venue_tag,academic, deepjlu_graph,463,Research Papers > System Intelligence > State Management,Living-Harness Is an Interactive-Agent Evolver,https://arxiv.org/abs/2607.26598,2607.26598,arXiv,2026,7,arxiv_id,academic, deepjlu_graph,464,Research Papers > System Intelligence > State Management,APEX: Autonomous Policy Exploration for Self-Evolving LLM Agents,https://arxiv.org/abs/2605.21240,2605.21240,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,468,Research Papers > System Intelligence > System Evolution,QueenBee Planner: Skill-Evolving Communication Topologies for Token-Efficient LLM Multi-Agent Systems,https://arxiv.org/abs/2606.27492,2606.27492,arXiv,2026,6,arxiv_id,academic, deepjlu_graph,469,Research Papers > System Intelligence > System Evolution,Recreate: Reasoning and creating domain agents driven by experience,https://aclanthology.org/2026.acl-long.1432/,,ACL,2026,,venue_tag,academic, deepjlu_graph,470,Research Papers > System Intelligence > System Evolution,SkillGraph: Self-Evolving Multi-Agent Collaboration with Multimodal Graph Topology,https://arxiv.org/abs/2604.17503,2604.17503,arXiv,2026,4,arxiv_id,academic, deepjlu_graph,471,Research Papers > System Intelligence > System Evolution,"Swarm Skills: A Portable, Self-Evolving Multi-Agent System Specification for Coordination Engineering",https://arxiv.org/abs/2605.10052,2605.10052,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,472,Research Papers > System Intelligence > System Evolution,MemTX: Transactional Belief Commit for Stateful Agent Memory,https://arxiv.org/abs/2607.23929,2607.23929,arXiv,2026,7,arxiv_id,academic, deepjlu_graph,473,Research Papers > System Intelligence > System Evolution,"The Log is the Agent: Event-Sourced Reactive Graphs for Auditable, Forkable Agentic Systems",https://arxiv.org/abs/2605.21997,2605.21997,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,474,Research Papers > System Intelligence > System Evolution,TDAG: A Multi-Agent Framework Based on Dynamic Task Decomposition and Agent Generation,https://www.sciencedirect.com/science/article/abs/pii/S0893608025000796,,Neural Networks,2025,,venue_tag,academic, deepjlu_graph,475,Research Papers > System Intelligence > System Evolution,Flow: Modularized Agentic Workflow Automation,https://proceedings.iclr.cc/paper_files/paper/2025/hash/ba84da6921f3040b74ee163aa7451f53-Abstract-Conference.html,,ICLR,2025,,venue_tag,academic, deepjlu_graph,476,Research Papers > System Intelligence > System Evolution,DynTaskMAS: A dynamic task graph-driven framework for asynchronous and parallel llm-based multi-agent systems,https://scholar.google.com/scholar?q=Dyntaskmas%3A+A+dynamic+task+graph-driven+framework+for+asynchronous+and+parallel+llm-based+multi-agent+systems,,ICAPS,2025,,venue_tag,academic, deepjlu_graph,477,Research Papers > System Intelligence > System Evolution,DyFlow: Dynamic Workflow Framework for Agentic Reasoning,https://proceedings.neurips.cc/paper_files/paper/2025/hash/fe9910d2b03324faeb5371a9658277bb-Abstract-Conference.html,,NeurIPS,2025,,venue_tag,academic, deepjlu_graph,478,Research Papers > System Intelligence > System Evolution,EvoFlow: Evolving Diverse Agentic Workflows On The Fly,https://arxiv.org/abs/2502.07373,2502.07373,arXiv,2025,2,arxiv_id,academic, deepjlu_graph,479,Research Papers > System Intelligence > System Evolution,QualityFlow: An Agentic Workflow for Program Synthesis Controlled by LLM Quality Checks,https://arxiv.org/abs/2501.17167,2501.17167,arXiv,2025,1,arxiv_id,academic, deepjlu_graph,480,Research Papers > System Intelligence > System Evolution,FlowSteer: Prompt-Only Workflow Steering Exposes Planning-Time Vulnerabilities in Multi-Agent LLM Systems,https://arxiv.org/abs/2605.11514,2605.11514,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,481,Research Papers > System Intelligence > System Evolution,SwarmAgentic: Towards fully automated agentic system generation via swarm intelligence,https://aclanthology.org/2025.emnlp-main.93/,,EMNLP,2025,,venue_tag,academic, deepjlu_graph,482,Research Papers > System Intelligence > System Evolution,AgentNet: Decentralized evolutionary coordination for llm-based multi-agent systems,https://proceedings.neurips.cc/paper_files/paper/2025/hash/9a379c1b05793d1c42dc832269834515-Abstract-Conference.html,,NeurIPS,2025,,venue_tag,academic, deepjlu_graph,483,Research Papers > System Intelligence > System Evolution,Drop the Hierarchy and Roles: How Self-Organizing LLM Agents Outperform Designed Structures,https://arxiv.org/abs/2603.28990,2603.28990,arXiv,2026,3,arxiv_id,academic, deepjlu_graph,484,Research Papers > System Intelligence > System Evolution,Evolve as a Team: Collaborative Self-Evolution for LLM-based Multi-Agent Systems,https://arxiv.org/abs/2605.29790,2605.29790,arXiv,2026,5,arxiv_id,academic, deepjlu_graph,485,Research Papers > System Intelligence > System Evolution,DyTopo: Dynamic topology routing for multi-agent reasoning via semantic matching,https://arxiv.org/abs/2602.06039,2602.06039,arXiv,2026,2,arxiv_id,academic, deepjlu_graph,486,Research Papers > System Intelligence > System Evolution,CARD: Towards Conditional Design of Multi-agent Topological Structures,https://arxiv.org/abs/2603.01089,2603.01089,arXiv,2026,3,arxiv_id,academic, deepjlu_graph,491,Research Papers > System Intelligence > Ontology Engineering,OntoExtend: A Framework for Requirement-driven and Scalable Ontology Extension with LLMs,https://arxiv.org/abs/2607.17963,2607.17963,arXiv,2026,7,arxiv_id,academic, deepjlu_graph,493,Research Papers > System Intelligence > Ontology Engineering,iCARE: Ontology-Guided Intent Routing for Multi-Agent LLM-Based Dialogue Systems,https://scholar.google.com/scholar?q=iCARE%3A+Ontology-Guided+Intent+Routing+for+Multi-Agent+LLM-Based+Dialogue+Systems,,Paper,2025,,venue_tag,academic, deepjlu_graph,494,Research Papers > System Intelligence > Ontology Engineering,Toward Culturally Aligned LLMs through Ontology-Guided Multi-Agent Reasoning,https://arxiv.org/abs/2601.21700,2601.21700,arXiv,2026,1,arxiv_id,academic, deepjlu_graph,495,Research Papers > System Intelligence > Ontology Engineering,Agentic Information Architectures for Global Climate Governance: A Multi-Agent Decision-Support System for Cross-National Policy Analytics,https://scholar.google.com/scholar?q=Agentic+Information+Architectures+for+Global+Climate+Governance%3A+A+Multi-Agent+Decision-Support+System+for+Cross-National+Policy+Analytics,,Paper,2026,,venue_tag,academic, deepjlu_graph,496,Research Papers > System Intelligence > Ontology Engineering,LaMAS4PD-A Multi-Agent LLM Approach for Ontology-Driven Structuring of Engineering Knowledge in Industry 4.0,https://scholar.google.com/scholar?q=LaMAS4PD-A+Multi-Agent+LLM+Approach+for+Ontology-Driven+Structuring+of+Engineering+Knowledge+in+Industry+4.0,,Procedia CIRP,2026,,venue_tag,academic, deepjlu_graph,497,Research Papers > System Intelligence > Ontology Engineering,Agentology: Ontology-Driven Operational Environments for Multi-Agent Systems,https://scholar.google.com/scholar?q=Agentology%3A+Ontology-Driven+Operational+Environments+for+Multi-Agent+Systems,,Available at SSRN 6919461,2026,,venue_tag,academic, deepjlu_graph,498,Research Papers > System Intelligence > Ontology Engineering,OntoCodex: a multi-agent biomedical ontology enrichment framework,https://scholar.google.com/scholar?q=OntoCodex%3A+a+multi-agent+biomedical+ontology+enrichment+framework,,npj Health Systems,2026,,venue_tag,academic, deepjlu_graph,499,Research Papers > System Intelligence > Ontology Engineering,AgentO: An Ontology for Modeling Agentic AI Systems,https://scholar.google.com/scholar?q=AgentO%3A+An+Ontology+for+Modeling+Agentic+AI+Systems,,European Semantic Web Conference,2026,,venue_tag,academic, deepjlu_graph,500,Research Papers > System Intelligence > Ontology Engineering,Ontology-to-tools compilation for executable semantic constraint enforcement in LLM agents,https://arxiv.org/abs/2602.03439,2602.03439,arXiv,2026,2,arxiv_id,academic, deepjlu_graph,501,Research Papers > System Intelligence > Ontology Engineering,Large language models for ontology engineering: a systematic literature review,https://scholar.google.com/scholar?q=Large+language+models+for+ontology+engineering%3A+a+systematic+literature+review,,Semantic Web,2026,,venue_tag,academic, deepjlu_graph,502,Research Papers > System Intelligence > Ontology Engineering,The Ontology System,https://www.palantir.com/docs/foundry/architecture-center/ontology-system,,arXiv,2026,,venue_tag,grey, deepjlu_graph,511,"Benchmarks, Datasets, and Environments > Model Intelligence",MMLU/MMLU-Pro,https://arxiv.org/abs/2406.01574,2406.01574,,2024,6,arxiv_id,academic, deepjlu_graph,512,"Benchmarks, Datasets, and Environments > Model Intelligence",GPQA,https://arxiv.org/abs/2311.12022,2311.12022,,2023,11,arxiv_id,academic, deepjlu_graph,513,"Benchmarks, Datasets, and Environments > Model Intelligence",NPPC,https://arxiv.org/abs/2504.11239,2504.11239,,2025,4,arxiv_id,academic, deepjlu_graph,514,"Benchmarks, Datasets, and Environments > Model Intelligence",OlymMATH,https://aclanthology.org/2026.acl-long.792/,,,,,,academic, deepjlu_graph,515,"Benchmarks, Datasets, and Environments > Model Intelligence",IFEval,https://arxiv.org/abs/2311.07911,2311.07911,,2023,11,arxiv_id,academic, deepjlu_graph,516,"Benchmarks, Datasets, and Environments > Model Intelligence",EvolIF,https://aclanthology.org/2026.acl-long.433/,,,,,,academic, deepjlu_graph,517,"Benchmarks, Datasets, and Environments > Model Intelligence",HumanEval/EvalPlus,https://arxiv.org/abs/2305.01210,2305.01210,,2023,5,arxiv_id,academic, deepjlu_graph,518,"Benchmarks, Datasets, and Environments > Model Intelligence",MMMU,https://openaccess.thecvf.com/content/CVPR2024/html/Yue_MMMU_A_Massive_Multi-discipline_Multimodal_Understanding_and_Reasoning_Benchmark_for_CVPR_2024_paper.html,,,,,,academic, deepjlu_graph,519,"Benchmarks, Datasets, and Environments > Model Intelligence",OMHBench,https://aclanthology.org/2026.findings-acl.911/,,,,,,academic, deepjlu_graph,520,"Benchmarks, Datasets, and Environments > Model Intelligence",LiveBench,https://arxiv.org/abs/2406.19314,2406.19314,,2024,6,arxiv_id,academic, deepjlu_graph,521,"Benchmarks, Datasets, and Environments > Model Intelligence",GraphRAG-Bench,https://proceedings.iclr.cc/paper_files/paper/2026/hash/6c9e01d6cefbbf4cdd265032550e767f-Abstract-Conference.html,,,,,,academic, deepjlu_graph,525,"Benchmarks, Datasets, and Environments > Individual Intelligence",AgentBench,https://arxiv.org/abs/2308.03688,2308.03688,,2023,8,arxiv_id,academic, deepjlu_graph,526,"Benchmarks, Datasets, and Environments > Individual Intelligence",GAIA,https://arxiv.org/abs/2311.12983,2311.12983,,2023,11,arxiv_id,academic, deepjlu_graph,527,"Benchmarks, Datasets, and Environments > Individual Intelligence",AgencyBench,https://aclanthology.org/2026.acl-long.337/,,,,,,academic, deepjlu_graph,528,"Benchmarks, Datasets, and Environments > Individual Intelligence",WebArena,https://arxiv.org/abs/2307.13854,2307.13854,,2023,7,arxiv_id,academic, deepjlu_graph,529,"Benchmarks, Datasets, and Environments > Individual Intelligence",OSWorld,https://arxiv.org/abs/2404.07972,2404.07972,,2024,4,arxiv_id,academic, deepjlu_graph,530,"Benchmarks, Datasets, and Environments > Individual Intelligence",SWE-bench,https://arxiv.org/abs/2310.06770,2310.06770,,2023,10,arxiv_id,academic, deepjlu_graph,531,"Benchmarks, Datasets, and Environments > Individual Intelligence",AppWorld,https://appworld.dev/,,,,,,grey, deepjlu_graph,532,"Benchmarks, Datasets, and Environments > Individual Intelligence",$$-bench,https://arxiv.org/abs/2406.12045,2406.12045,,2024,6,arxiv_id,academic, deepjlu_graph,533,"Benchmarks, Datasets, and Environments > Individual Intelligence",ToolSandbox,https://github.com/apple/ToolSandbox,,,,,,repository, deepjlu_graph,534,"Benchmarks, Datasets, and Environments > Individual Intelligence",AgentDojo,https://github.com/ethz-spylab/agentdojo,,,,,,repository, deepjlu_graph,535,"Benchmarks, Datasets, and Environments > Individual Intelligence",Harness-Bench,https://arxiv.org/abs/2605.27922,2605.27922,,2026,5,arxiv_id,academic, deepjlu_graph,536,"Benchmarks, Datasets, and Environments > Individual Intelligence",Skill-Use,https://arxiv.org/abs/2608.04828,2608.04828,,2026,8,arxiv_id,academic, deepjlu_graph,537,"Benchmarks, Datasets, and Environments > Individual Intelligence",LongMemEval,https://arxiv.org/abs/2410.10813,2410.10813,,2024,10,arxiv_id,academic, deepjlu_graph,538,"Benchmarks, Datasets, and Environments > Individual Intelligence",MemoryAgentBench,https://github.com/HUST-AI-HYZ/MemoryAgentBench,,,,,,repository, deepjlu_graph,539,"Benchmarks, Datasets, and Environments > Individual Intelligence",MemoryArena,https://arxiv.org/abs/2602.16313,2602.16313,,2026,2,arxiv_id,academic, deepjlu_graph,540,"Benchmarks, Datasets, and Environments > Individual Intelligence",GateMem,https://rzhub.github.io/GateMem/project.html,,,,,,grey, deepjlu_graph,541,"Benchmarks, Datasets, and Environments > Individual Intelligence",MemSyco-Bench,https://arxiv.org/abs/2607.01071,2607.01071,,2026,7,arxiv_id,academic, deepjlu_graph,542,"Benchmarks, Datasets, and Environments > Individual Intelligence",Mem2ActBench,https://github.com/Cantaloupe-M/Mem2ActBench,,,,,,repository, deepjlu_graph,543,"Benchmarks, Datasets, and Environments > Individual Intelligence",LongDS-Bench,https://arxiv.org/abs/2605.30434,2605.30434,,2026,5,arxiv_id,academic, deepjlu_graph,544,"Benchmarks, Datasets, and Environments > Individual Intelligence",EvoMemBench,https://github.com/DSAIL-Memory/EvoMemBench,,,,,,repository, deepjlu_graph,545,"Benchmarks, Datasets, and Environments > Individual Intelligence",Trainee-Bench,https://arxiv.org/abs/2601.08173,2601.08173,,2026,1,arxiv_id,academic, deepjlu_graph,546,"Benchmarks, Datasets, and Environments > Individual Intelligence",SEA-Eval,https://arxiv.org/abs/2604.08988,2604.08988,,2026,4,arxiv_id,academic, deepjlu_graph,547,"Benchmarks, Datasets, and Environments > Individual Intelligence",Evo-Bench,https://arxiv.org/abs/2608.09096,2608.09096,,2026,8,arxiv_id,academic, deepjlu_graph,548,"Benchmarks, Datasets, and Environments > Individual Intelligence",OpenClawBench,https://arxiv.org/abs/2605.29253,2605.29253,,2026,5,arxiv_id,academic, deepjlu_graph,549,"Benchmarks, Datasets, and Environments > Individual Intelligence",BenchTrace,https://arxiv.org/abs/2605.29225,2605.29225,,2026,5,arxiv_id,academic, deepjlu_graph,550,"Benchmarks, Datasets, and Environments > Individual Intelligence",TheAgentCompany,https://arxiv.org/abs/2412.14161,2412.14161,,2024,12,arxiv_id,academic, deepjlu_graph,554,"Benchmarks, Datasets, and Environments > System Intelligence",TaskBench,https://arxiv.org/abs/2311.18760,2311.18760,,2023,11,arxiv_id,academic, deepjlu_graph,555,"Benchmarks, Datasets, and Environments > System Intelligence",WorFBench,https://arxiv.org/abs/2410.07869,2410.07869,,2024,10,arxiv_id,academic, deepjlu_graph,556,"Benchmarks, Datasets, and Environments > System Intelligence",FlowBench,https://aclanthology.org/2024.findings-emnlp.638/,,,,,,academic, deepjlu_graph,557,"Benchmarks, Datasets, and Environments > System Intelligence",ComfyBench,https://openaccess.thecvf.com/content/CVPR2025/html/Xue_ComfyBench_Benchmarking_LLM-based_Agents_in_ComfyUI_for_Autonomously_Designing_Collaborative_CVPR_2025_paper.html,,,,,,academic, deepjlu_graph,558,"Benchmarks, Datasets, and Environments > System Intelligence",TPS-Bench,https://aclanthology.org/2026.acl-long.1614/,,,,,,academic, deepjlu_graph,559,"Benchmarks, Datasets, and Environments > System Intelligence",JourneyBench,https://aclanthology.org/2026.eacl-industry.15/,,,,,,academic, deepjlu_graph,560,"Benchmarks, Datasets, and Environments > System Intelligence",ETOM,https://aclanthology.org/2026.findings-eacl.75/,,,,,,academic, deepjlu_graph,561,"Benchmarks, Datasets, and Environments > System Intelligence",LLM-Coordination,https://aclanthology.org/2025.findings-naacl.448/,,,,,,academic, deepjlu_graph,562,"Benchmarks, Datasets, and Environments > System Intelligence",VillagerBench,https://aclanthology.org/2024.findings-acl.964/,,,,,,academic, deepjlu_graph,563,"Benchmarks, Datasets, and Environments > System Intelligence",MultiAgentBench,https://aclanthology.org/2025.acl-long.421/,,,,,,academic, deepjlu_graph,564,"Benchmarks, Datasets, and Environments > System Intelligence",AgentsNet,https://arxiv.org/abs/2507.08616,2507.08616,,2025,7,arxiv_id,academic, deepjlu_graph,566,"Benchmarks, Datasets, and Environments > System Intelligence",SILO-BENCH,https://aclanthology.org/2026.acl-long.1354/,,,,,,academic, deepjlu_graph,567,"Benchmarks, Datasets, and Environments > System Intelligence",CoLLAB,https://openreview.net/forum?id=372FjQy1cF,,,,,,academic, deepjlu_graph,568,"Benchmarks, Datasets, and Environments > System Intelligence",Collab-Overcooked,https://aclanthology.org/2025.emnlp-main.249/,,,,,,academic, deepjlu_graph,569,"Benchmarks, Datasets, and Environments > System Intelligence",MAS-BENCH,https://aclanthology.org/2026.findings-acl.1698/,,,,,,academic, deepjlu_graph,570,"Benchmarks, Datasets, and Environments > System Intelligence",DPBench,https://github.com/najmulhasan-code/dpbench,,,,,,repository, deepjlu_graph,571,"Benchmarks, Datasets, and Environments > System Intelligence",CalBench,https://arxiv.org/abs/2605.09823,2605.09823,,2026,5,arxiv_id,academic, deepjlu_graph,572,"Benchmarks, Datasets, and Environments > System Intelligence",TAMAS,https://aclanthology.org/2026.acl-long.1442/,,,,,,academic, deepjlu_graph,573,"Benchmarks, Datasets, and Environments > System Intelligence",SyncBench,https://proceedings.mlr.press/v267/guo25l.html,,,,,,academic, deepjlu_graph,574,"Benchmarks, Datasets, and Environments > System Intelligence",MAST,https://arxiv.org/abs/2503.13657,2503.13657,,2025,3,arxiv_id,academic, deepjlu_graph,575,"Benchmarks, Datasets, and Environments > System Intelligence",TraceElephant,https://aclanthology.org/2026.acl-long.912/,,,,,,academic, deepjlu_graph,576,"Benchmarks, Datasets, and Environments > System Intelligence",MP-Bench,https://arxiv.org/abs/2603.25001,2603.25001,,2026,3,arxiv_id,academic, deepjlu_graph,577,"Benchmarks, Datasets, and Environments > System Intelligence",R2Act,https://arxiv.org/abs/2607.04623,2607.04623,,2026,7,arxiv_id,academic, deepjlu_graph,578,"Benchmarks, Datasets, and Environments > System Intelligence",MAFBench,https://arxiv.org/abs/2602.03128,2602.03128,,2026,2,arxiv_id,academic, deepjlu_graph,579,"Benchmarks, Datasets, and Environments > System Intelligence",MASEval,https://aclanthology.org/2026.acl-demo.34/,,,,,,academic, deepjlu_graph,580,"Benchmarks, Datasets, and Environments > System Intelligence",MAS-PromptBench,https://arxiv.org/abs/2606.23664,2606.23664,,2026,6,arxiv_id,academic, deepjlu_graph,581,"Benchmarks, Datasets, and Environments > System Intelligence",BenchAgent,https://arxiv.org/abs/2606.05670,2606.05670,,2026,6,arxiv_id,academic, deepjlu_graph,582,"Benchmarks, Datasets, and Environments > System Intelligence",PerspectiveGap,https://arxiv.org/abs/2606.08878,2606.08878,,2026,6,arxiv_id,academic, deepjlu_graph,590,Open-Source Libraries > Model Intelligence,Transformers,https://github.com/huggingface/transformers,,,,,,repository, deepjlu_graph,591,Open-Source Libraries > Model Intelligence,Megatron Core,https://github.com/NVIDIA/Megatron-LM,,,,,,repository, deepjlu_graph,592,Open-Source Libraries > Model Intelligence,LLaMA-Factory,https://github.com/hiyouga/LlamaFactory,,,,,,repository, deepjlu_graph,593,Open-Source Libraries > Model Intelligence,verl,https://github.com/verl-project/verl,,,,,,repository, deepjlu_graph,594,Open-Source Libraries > Model Intelligence,slime,https://github.com/THUDM/slime,,,,,,repository, deepjlu_graph,595,Open-Source Libraries > Model Intelligence,vLLM,https://github.com/vllm-project/vllm,,,,,,repository, deepjlu_graph,596,Open-Source Libraries > Model Intelligence,SGLang,https://github.com/sgl-project/sglang,,,,,,repository, deepjlu_graph,597,Open-Source Libraries > Model Intelligence,Reef,https://github.com/Human-Agent-Society/reef,,,,,,repository, deepjlu_graph,601,Open-Source Libraries > Individual Intelligence,LangChain,https://github.com/langchain-ai/langchain,,,,,,repository, deepjlu_graph,602,Open-Source Libraries > Individual Intelligence,OpenAI Agents SDK,https://github.com/openai/openai-agents-python,,,,,,repository, deepjlu_graph,603,Open-Source Libraries > Individual Intelligence,Claude Agent SDK,https://github.com/anthropics/claude-agent-sdk-python,,,,,,repository, deepjlu_graph,604,Open-Source Libraries > Individual Intelligence,Pydantic AI,https://github.com/pydantic/pydantic-ai,,,,,,repository, deepjlu_graph,605,Open-Source Libraries > Individual Intelligence,LlamaIndex Workflows,https://github.com/run-llama/workflows-py,,,,,,repository, deepjlu_graph,606,Open-Source Libraries > Individual Intelligence,Haystack,https://github.com/deepset-ai/haystack,,,,,,repository, deepjlu_graph,607,Open-Source Libraries > Individual Intelligence,Apache Burr,https://github.com/apache/burr,,,,,,repository, deepjlu_graph,608,Open-Source Libraries > Individual Intelligence,Letta Agent SDK,https://github.com/letta-ai/letta-agent-sdk,,,,,,repository, deepjlu_graph,609,Open-Source Libraries > Individual Intelligence,Graphiti,https://github.com/getzep/graphiti,,,,,,repository, deepjlu_graph,610,Open-Source Libraries > Individual Intelligence,MCP Python SDK,https://github.com/modelcontextprotocol/python-sdk,,,,,,repository, deepjlu_graph,611,Open-Source Libraries > Individual Intelligence,Langflow,https://github.com/langflow-ai/langflow,,,,,,repository, deepjlu_graph,612,Open-Source Libraries > Individual Intelligence,Dify,https://github.com/langgenius/dify,,,,,,repository, deepjlu_graph,618,Open-Source Libraries > System Intelligence > Graph Engineering,LangGraph,https://github.com/langchain-ai/langgraph,,,,,,repository, deepjlu_graph,619,Open-Source Libraries > System Intelligence > Graph Engineering,Microsoft Agent Framework,https://github.com/microsoft/agent-framework,,,,,,repository, deepjlu_graph,620,Open-Source Libraries > System Intelligence > Graph Engineering,Google ADK,https://github.com/google/adk-python,,,,,,repository, deepjlu_graph,621,Open-Source Libraries > System Intelligence > Graph Engineering,AutoGen/GraphFlow,https://github.com/microsoft/autogen,,,,,,repository, deepjlu_graph,623,Open-Source Libraries > System Intelligence > Graph Engineering,CrewAI,https://github.com/crewAIInc/crewAI,,,,,,repository, deepjlu_graph,624,Open-Source Libraries > System Intelligence > Graph Engineering,CAMEL,https://github.com/camel-ai/camel,,,,,,repository, deepjlu_graph,625,Open-Source Libraries > System Intelligence > Graph Engineering,Mastra,https://github.com/mastra-ai/mastra,,,,,,repository, deepjlu_graph,626,Open-Source Libraries > System Intelligence > Graph Engineering,GPTSwarm,https://github.com/metauto-ai/GPTSwarm,,,,,,repository, deepjlu_graph,627,Open-Source Libraries > System Intelligence > Graph Engineering,Raven,https://github.com/EverMind-AI/Raven,,,,,,repository, deepjlu_graph,631,Open-Source Libraries > System Intelligence > Ontology Engineering,Semantica,https://github.com/semantica-agi/semantica,,,,,,repository, deepjlu_graph,639,Applications > Software Engineering and IT Operations,MetaGPT,https://proceedings.iclr.cc/paper_files/paper/2024/hash/6507b115562bb0a305f1958ccc87355a-Abstract-Conference.html,,,,,,academic, deepjlu_graph,640,Applications > Software Engineering and IT Operations,SWE-agent,https://proceedings.neurips.cc/paper_files/paper/2024/hash/5a7c947568c1b1328ccc5230172e1e7c-Abstract-Conference.html,,,,,,academic, deepjlu_graph,641,Applications > Software Engineering and IT Operations,OpenHands,https://openreview.net/forum?id=OJd3ayDDoF,,,,,,academic, deepjlu_graph,642,Applications > Software Engineering and IT Operations,Codex,https://openai.com/index/introducing-the-codex-app/,,,,,,grey, deepjlu_graph,643,Applications > Software Engineering and IT Operations,Claude Code,https://www.anthropic.com/product/claude-code,,,,,,grey, deepjlu_graph,644,Applications > Software Engineering and IT Operations,OpenCode,https://opencode.ai/docs/agents/,,,,,,grey, deepjlu_graph,645,Applications > Software Engineering and IT Operations,Cline,https://docs.cline.bot/sdk/guides/multi-agent-teams,,,,,,grey, deepjlu_graph,646,Applications > Software Engineering and IT Operations,Project ALICE,https://research.ibm.com/blog/project-alice-software-bugs-agents,,,,,,grey, deepjlu_graph,650,Applications > Scientific Discovery and Laboratory Automation,SciAgents,https://doi.org/10.1002/adma.202413523,,,,,,academic, deepjlu_graph,651,Applications > Scientific Discovery and Laboratory Automation,The AI Scientist,https://github.com/SakanaAI/AI-Scientist,,,,,,repository, deepjlu_graph,652,Applications > Scientific Discovery and Laboratory Automation,Virtual Lab,https://doi.org/10.1038/s41586-025-09442-9,,,,,,academic, deepjlu_graph,653,Applications > Scientific Discovery and Laboratory Automation,Co-Scientist,https://doi.org/10.1038/s41586-026-10644-y,,,,,,academic, deepjlu_graph,654,Applications > Scientific Discovery and Laboratory Automation,Robin,https://doi.org/10.1038/s41586-026-10652-y,,,,,,academic, deepjlu_graph,655,Applications > Scientific Discovery and Laboratory Automation,Agon,https://github.com/AutoResearch-Factory/Agon,,,,,,repository, deepjlu_graph,656,Applications > Scientific Discovery and Laboratory Automation,AutoNumerics,https://arxiv.org/abs/2602.17607,2602.17607,,2026,2,arxiv_id,academic, deepjlu_graph,657,Applications > Scientific Discovery and Laboratory Automation,OptimAI,https://arxiv.org/abs/2504.16918,2504.16918,,2025,4,arxiv_id,academic, deepjlu_graph,661,Applications > Healthcare and Clinical Decision Support,DeepRare,https://doi.org/10.1038/s41586-025-10097-9,,,,,,academic, deepjlu_graph,662,Applications > Healthcare and Clinical Decision Support,AMIE,https://doi.org/10.1038/s41586-026-10764-5,,,,,,academic, deepjlu_graph,663,Applications > Healthcare and Clinical Decision Support,CARE-AD,https://doi.org/10.1038/s41746-025-01940-4,,,,,,academic, deepjlu_graph,668,Applications > Enterprise Workflows and Digital Organizations,WorkTeam,https://aclanthology.org/2025.naacl-industry.3/,,,,,,academic, deepjlu_graph,669,Applications > Enterprise Workflows and Digital Organizations,SOAN,https://arxiv.org/abs/2508.13732,2508.13732,,2025,8,arxiv_id,academic, deepjlu_graph,670,Applications > Enterprise Workflows and Digital Organizations,FinRobot-ERP,https://arxiv.org/abs/2506.01423,2506.01423,,2025,6,arxiv_id,academic, deepjlu_graph,671,Applications > Enterprise Workflows and Digital Organizations,Agent-Ops,https://aclanthology.org/2026.acl-industry.29/,,,,,,academic, deepjlu_graph,672,Applications > Enterprise Workflows and Digital Organizations,Gemini Enterprise Agentic RAG,https://research.google/blog/unlocking-dependable-responses-with-gemini-enterprise-agent-platforms-agentic-rag/,,,,,,grey, deepjlu_graph,676,Applications > General-Purpose Digital Agents and Personal Automation,OpenClaw,https://github.com/openclaw/openclaw,,,,,,repository, deepjlu_graph,677,Applications > General-Purpose Digital Agents and Personal Automation,Hermes Agent,https://github.com/NousResearch/hermes-agent,,,,,,repository, deepjlu_graph,681,Applications > Social and Economic Simulation,AgentSociety,https://aclanthology.org/2025.acl-industry.94/,,,,,,academic, deepjlu_graph,682,Applications > Social and Economic Simulation,EconAgent,https://aclanthology.org/2024.acl-long.829/,,,,,,academic, deepjlu_graph,683,Applications > Social and Economic Simulation,SRAP-Agent,https://aclanthology.org/2024.findings-emnlp.15/,,,,,,academic, deepjlu_graph,684,Applications > Social and Economic Simulation,TwinMarket,https://proceedings.neurips.cc/paper_files/paper/2025/hash/5bf234ecf83cd77bc5b77a24ba9338b0-Abstract-Conference.html,,,,,,academic, aiboost_harness,68,Foundations,My AI Adoption Journey,https://mitchellh.com/writing/my-ai-adoption-journey,,,2026,2,text_date,grey,"The essay that coined ""harness engineering"" (February 2026): Mitchell Hashimoto's six-stage adoption path culminates in engineering the harness itself — every agent mistake becomes a permanent fix, whether an AGENTS.md line or a purpose-built tool, so the error class never recurs. The Ghostty numbers (commits needing correction dropping from 40% to 3% over three months of accumulated harness fixes" aiboost_harness,69,Foundations,Harness Engineering,https://openai.com/index/harness-engineering/,,,,,,grey,OpenAI's framing of harness engineering as a discipline: how to design the scaffolding that lets Codex and similar agents operate reliably in an agent-first world. aiboost_harness,70,Foundations,Unrolling the Codex Agent Loop,https://openai.com/index/unrolling-the-codex-agent-loop/,,,,,,grey,"OpenAI's detailed breakdown of the Codex agent loop, exposing each harness component and where it can be improved." aiboost_harness,71,Foundations,Run Long-Horizon Tasks with Codex,https://developers.openai.com/blog/run-long-horizon-tasks-with-codex/,,,,,,grey,"OpenAI's practice guide for long-horizon task planning: introduces Plan.md, Implement.md, Documentation.md as reusable harness artifacts." aiboost_harness,72,Foundations,Building Effective Agents,https://www.anthropic.com/research/building-effective-agents,,,,,,grey,"Anthropic's foundational guide on agent architecture, covering when to use workflows vs. agents and how to compose primitives." aiboost_harness,73,Foundations,Harness Design for Long-Running Application Development,https://www.anthropic.com/engineering/harness-design-long-running-apps,,,,,,grey,"Anthropic's engineering blog on designing harnesses for sustained, multi-session development tasks. Key insight: every harness component assumes the model can't do something; those assumptions expire." aiboost_harness,74,Foundations,Writing Effective Tools for Agents,https://www.anthropic.com/engineering/writing-tools-for-agents,,,,,,grey,"Anthropic's guide on tool interface design: naming, schemas, error surfaces, and the principle that tool design is agent UX." aiboost_harness,75,Foundations,Beyond Permission Prompts,https://www.anthropic.com/engineering/claude-code-sandboxing,,,,,,grey,Anthropic on building structured permission and authorization systems into agent harnesses instead of relying on natural-language permission text. aiboost_harness,76,Foundations,Demystifying Evals for AI Agents,https://www.anthropic.com/engineering/demystifying-evals-for-ai-agents,,,,,,grey,"Anthropic's framework for evaluating agent behavior: what to measure, how to build eval harnesses, and why unit-test-style evals fail for agents." aiboost_harness,77,Foundations,What is an AI Agent?,https://www.ibm.com/think/topics/ai-agents,,,,,,grey,"IBM's definitional piece, useful for anchoring harness design decisions to a clear model of what an agent actually is." aiboost_harness,78,Foundations,Agent Development Kit: Making it easy to build multi-agent applications,https://developers.googleblog.com/en/agent-development-kit-easy-to-build-multi-agent-applications/,,,,,,grey,"Google's announcement and design rationale for ADK: explains the multi-agent topology, tool registration model, and eval pipeline that shaped their framework. Complements the Anthropic/OpenAI framing with Google's production perspective." aiboost_harness,79,Foundations,Harness Engineering,https://martinfowler.com/articles/exploring-gen-ai/harness-engineering.html,,,,,,grey,"Martin Fowler's synthesis of what harness engineering practice looks like: three interlocking systems — context engineering (curating what the agent knows), architectural constraints (deterministic linters and structural tests), and entropy management (periodic agents that repair documentation drift). The ""humans on the loop"" framing — harness engineers who design and maintain agent environments r" aiboost_harness,80,Foundations,The Anatomy of an Agent Harness,https://blog.langchain.com/the-anatomy-of-an-agent-harness/,,,,,,grey,"LangChain's structural breakdown of the five primitives that compose a harness: filesystem (durable state + agent collaboration surface), code execution (autonomous problem-solving without pre-designed solutions), sandbox (isolation + verification), memory (cross-session persistence), and context management (compaction against ""context rot""). The co-evolution warning — models trained with specific" aiboost_harness,81,Foundations,"Building AI Coding Agents for the Terminal: Scaffolding, Harness, Context Engineering, and Lessons Learned",https://arxiv.org/abs/2603.05344,2603.05344,,2026,3,arxiv_id,academic,"The first systematic practitioner paper on terminal-native coding agent harness design: eager-construction scaffolding (pre-build all components before the first message to eliminate first-call latency and race conditions), compound multi-model architecture (different model instances for execution, reasoning, critique, and vision tasks), 5-layer defense-in-depth safety, and schema-filtered plannin" aiboost_harness,82,Foundations,Natural-Language Agent Harnesses,https://arxiv.org/abs/2603.25723,2603.25723,,2026,3,arxiv_id,academic,"Proposes externalizing agent control logic as portable natural-language artifacts (NLAHs) executed by a shared Intelligent Harness Runtime, enabling harness design to be studied, transferred, and reproduced rather than buried in bespoke controller code. Directly addresses the root cause of harness fragility: control logic scattered across framework defaults and hard-coded controller logic that can" aiboost_harness,83,Foundations,Ranking Engineer Agent (REA): Meta's Autonomous AI System for Ads Ranking,https://engineering.fb.com/2026/03/17/developer-tools/ranking-engineer-agent-rea-autonomous-ai-system-accelerating-meta-ads-ranking-innovation/,,,2026,3,url_date,grey,Meta's Autonomous AI System for Ads Ranking](https://engineering.fb.com/2026/03/17/developer-tools/ranking-engineer-agent-rea-autonomous-ai-system-accelerating-meta-ads-ranking-innovation/) — Meta's production harness for multi-day ML pipeline automation with hibernate-and-wake checkpointing for resuming interrupted 6-hour tasks without losing context. Demonstrates harness design for scientific wo aiboost_harness,84,Foundations,Supercharge Your AI Agents: The New ADK Integrations Ecosystem,https://developers.googleblog.com/en/supercharge-your-ai-agents-adk-integrations-ecosystem/,,,,,,grey,"Google's 2026 update to Agent Development Kit expanding the ecosystem integrations (Hugging Face, GitHub, Daytona, Notion, etc.) and providing reference patterns for how orchestration harnesses wire external services without losing determinism or state coherence." aiboost_harness,85,Foundations,2026 Agentic Coding Trends Report,https://resources.anthropic.com/hubfs/2026%20Agentic%20Coding%20Trends%20Report.pdf?hsLang=en,,,,,,grey,"Anthropic's industry benchmark identifying infrastructure configuration as a first-class optimization variable: harness setup alone can swing benchmarks by 5+ percentage points. Documents the shift from single-agent to orchestrated multi-agent teams and introduces the ""agentic engineering platform"" category, bridging the gap between agent frameworks and production deployment infrastructure." aiboost_harness,86,Foundations,How We Build Azure SRE Agent with Agentic Workflows,https://techcommunity.microsoft.com/blog/appsonazureblog/how-we-build-azure-sre-agent-with-agentic-workflows/4508753,,,,,,grey,"Architecture walkthrough of Microsoft's agent that has handled 35,000+ production incidents autonomously, reducing Azure App Service time-to-mitigation from 40.5 hours to 3 minutes. Documents the integration of MCP tools, telemetry, code repositories, and incident management platforms into a single agent harness with human-in-the-loop governance. The most data-backed production harness case study " aiboost_harness,87,Foundations,Context Engineering for Reliable AI Agents: Lessons from Building Azure SRE Agent,https://techcommunity.microsoft.com/blog/appsonazureblog/context-engineering-lessons-from-building-azure-sre-agent/4481200/,,,,,,grey,"Microsoft's account of shifting from 100+ bespoke tools and a prescriptive prompt to a filesystem-based context engineering system for their SRE agent. Key finding: exposing everything (source code, runbooks, query schemas, past investigation notes) as files and letting the agent use read_file, grep, find, and shell outperformed specialized tooling — ""Intent Met"" score rose from 45% to 75% on nove" aiboost_harness,88,Foundations,Harness Engineering: Structured Workflows for AI-Assisted Development,https://developers.redhat.com/articles/2026/04/07/harness-engineering-structured-workflows-ai-assisted-development,,,2026,4,text_date,grey,"Red Hat's enterprise perspective on harness engineering (April 7, 2026): AI writes better code when you design the environment it works in. Emphasizes structured context over free-form tickets, expanding the agent's toolbox through MCP integrations (CI status, deployment logs, runtime metrics) as real data sources, and a four-pillar model (vibes, specs, skills, agents) for organizing how humans an" aiboost_harness,89,Foundations,Harness engineering for coding agent users,https://martinfowler.com/articles/harness-engineering.html,,,2026,4,text_date,grey,"Birgitta Böckeler's systematic mental model (April 2026) for coding-agent harnesses, framing them as feedforward guides plus feedback sensors that self-correct before output reaches human eyes. Distinguishes computational controls (linters, tests) from inferential ones (LLM-as-judge), and argues that harnessability should become a first-class criterion in technology and architecture decisions." aiboost_harness,90,Foundations,A Practical Guide to Building AI Agents,https://openai.com/business/guides-and-resources/a-practical-guide-to-building-ai-agents/,,,2026,4,text_date,grey,"OpenAI's April 2026 comprehensive guide distilling production deployment patterns into actionable best practices: single-agent vs. multi-agent orchestration (manager vs. decentralized handoffs), tool design for many-to-many agent-tool relationships, and layered guardrail patterns combining input validation, output filtering, tool-risk ratings, and human-intervention triggers." aiboost_harness,91,Foundations,An Update on Recent Claude Code Quality Reports,https://www.anthropic.com/engineering/april-23-postmortem,,,2026,4,text_date,grey,"Anthropic's transparent April 2026 postmortem tracing Claude Code quality degradation to three independent harness-level changes: a default reasoning-effort downgrade, a caching-optimization bug that continuously dropped thinking history from stale sessions, and an overly aggressive verbosity-limiting system prompt. Essential reading for understanding how seemingly minor harness adjustments — prom" aiboost_harness,92,Foundations,Agent Harness Design: 3 Patterns for Harnessing Claude's Intelligence,https://claude.com/blog/harnessing-claudes-intelligence,,,2026,4,text_date,grey,"Anthropic's April 2026 design guide distilling harness engineering into three actionable patterns: build on tools Claude already knows, remove harness assumptions as capabilities improve, and set UX/cost/safety boundaries carefully. A practical complement to the ""agent = model + harness"" framing that helps teams decide what scaffolding to keep, add, or remove over time." aiboost_harness,93,Foundations,Code as Agent Harness,https://arxiv.org/abs/2605.18747,2605.18747,,2026,5,arxiv_id,academic,"May 2026 survey framing code as the basis for agent infrastructure rather than merely output: it unifies harness interface, mechanisms, and multi-agent scaling through shared code artifacts, and surfaces open challenges from verification under incomplete feedback to regression-free improvement." aiboost_harness,94,Foundations,"Harness Engineering: How to Build Reliable AI Agents by Engineering the System, Not the Model",https://www.deepset.ai/blog/harness-engineering,,,2026,5,text_date,grey,"deepset's May 2026 synthesis of agent reliability as a harness problem: a failure-classification framework (context, constraint, verification, planning failures) that maps each failure mode to the right harness component, and a concrete demonstration that harness-only changes can move agents 20+ ranking positions without swapping the model." aiboost_harness,95,Foundations,What makes a harness a harness: necessary and sufficient conditions for an agent harness,https://arxiv.org/abs/2606.10106,2606.10106,,2026,6,arxiv_id,academic,"June 2026 constitutive definition of an agent harness as a runtime layer with four necessary and sufficient elements: an agent loop, a tool interface, context management, and control mechanisms. Applied to Claude Code, Codex CLI, Aider, Cline, OpenHands, and SWE-agent, it provides a rigorous inclusion test for distinguishing harnesses from generators, guardrails, or plain tool wrappers." aiboost_harness,96,Foundations,AI Harness Engineering: A Runtime Substrate for Foundation-Model Software Agents,https://arxiv.org/abs/2605.13357,2605.13357,,2026,5,arxiv_id,academic,"May 2026 formalization of the harness as the runtime substrate that mediates model–environment interaction: eleven component responsibilities (task specification, verification, failure attribution, and more) operationalized as a four-level ladder (H0–H3) that progressively exposes runtime support to the agent, plus a trace-based protocol that packages runs into auditable episode files. Complements" aiboost_harness,97,Foundations,Architectural Design Decisions in AI Agent Harnesses,https://arxiv.org/abs/2604.18071,2604.18071,,2026,4,arxiv_id,academic,"April 2026 empirical study of 70 public agent systems across five recurring dimensions (subagent architecture, context management, tool systems, safety mechanisms, orchestration) that synthesizes five architectural patterns. The comparative research package turns harness selection from a framework popularity contest into a reasoned comparison of design trade-offs." aiboost_harness,98,Foundations,RUCAIBox/awesome-agent-harness,https://github.com/RUCAIBox/awesome-agent-harness,,,,,,repository,"RUCAIBox's survey paper and curated reading list on *Agent Systems with Harness Engineering*, mapping harness design across agent workflows, memory systems, skill libraries, and multi-agent orchestration with 500+ references. The clearest academic complement to vendor-specific harness engineering posts." aiboost_harness,99,Foundations,"Tuning the harness, not the model: a Nemotron 3 Ultra playbook",https://blog.langchain.com/tuning-the-harness-not-the-model-a-nemotron-3-ultra-playbook,,,2026,7,text_date,grey,LangChain's July 2026 playbook showing how harness-only tuning brought Nemotron 3 Ultra within one point of Opus 4.8 on Deep Agents at roughly one-tenth the cost ($4.48 vs $43.48). The clearest recent demonstration that evals are the training data for harness work and that fit — not raw model capability — determines how much quality reaches the task. aiboost_harness,100,Foundations,lopopolo/harness-engineering,https://github.com/lopopolo/harness-engineering,,,,,,repository,"Ryan Lopopolo's anthology, field guide, and agent context bundle for harness engineering: it reframes the harness as the environment that carries an organization's nonfunctional requirements, with reusable AGENTS.md/CLAUDE.md artifacts, playbooks, evals, and domain modeling docs. The most systematic open-source synthesis of how to make organizational judgment cumulative across agent-maintained rep" aiboost_harness,101,Foundations,Harness Engineering for Self-Improvement,https://lilianweng.github.io/posts/2026-07-04-harness/,,,2026,7,text_date,grey,"Lilian Weng's July 2026 synthesis framing the harness as the deployment layer of recursive self-improvement: three design patterns (workflow automation, filesystem as persistent memory, sub-agents/backend jobs), a case study of the tool surface stabilized across Claude Code, Codex, OpenCode, and Cursor, and a map of self-improving-harness research from STOP through Meta-Harness, ADAS, AFlow, and A" aiboost_harness,102,Foundations,The State Of AI Harness Engineering 2026,https://marmelab.com/blog/2026/09/24/the-state-of-ai-harness-engineering-2026.html,,,2026,9,text_date,grey,"François Zaninotto's September 2026 synthesis grounded in an audit of 246 open-source harness repos and 57 publications: 60% of harnesses ship without tests or evals, machine-generated context files *reduce* success rates while increasing inference cost (ETH Zurich), more than four agent-to-agent handoffs almost always fail (Microsoft), and reviewer agents can degrade performance. The headline dat" aiboost_harness,103,Foundations,"Harness Engineering: Anatomy, Architecture, and Evolution of Coding Agents",https://arxiv.org/abs/2609.00006,2609.00006,,2026,9,arxiv_id,academic,"July 2026 source-code study of eleven production coding harnesses (Claude Code, Codex CLI, Gemini CLI, Mistral Vibe, OpenHands, Aider, Mini-SWE-Agent, Hermes, Pi, OpenCode, OpenClaw) with Omnigent as a meta-harness contrast point: seven canonical subsystems each mapped at their minimal and maximal implementation, yielding 13 cross-cutting observations and a catalog of 29 recurring design patterns." aiboost_harness,113,Design Primitives > Agent Loop,ReAct: Synergizing Reasoning and Acting in Language Models,https://arxiv.org/abs/2210.03629,2210.03629,,2022,10,arxiv_id,academic,The foundational paper defining the Thought/Action/Observation loop structure that underlies virtually every agent harness. Required reading for understanding why the loop is structured the way it is and where each harness component maps onto the reasoning-acting cycle. aiboost_harness,114,Design Primitives > Agent Loop,Unrolling the Codex Agent Loop,https://openai.com/index/unrolling-the-codex-agent-loop/,,,,,,grey,"The canonical decomposition of what happens inside one agent loop iteration: observe, plan, act, verify." aiboost_harness,115,Design Primitives > Agent Loop,LangGraph — Low Level Concepts,https://langchain-ai.github.io/langgraph/concepts/low_level/,,,,,,grey,"Models the agent loop explicitly as a directed graph with typed state, conditional edges, and checkpointing. The most concrete engineering treatment of loop control flow: how to implement termination conditions, branch on tool results, and persist mid-loop state for resumption." aiboost_harness,116,Design Primitives > Agent Loop,Unlocking the Codex Harness: How We Built the App Server,https://openai.com/index/unlocking-the-codex-harness/,,,,,,grey,"OpenAI's engineering deep-dive into the Item/Turn/Thread protocol (JSON-RPC/JSONL over stdio) that exposes the Codex harness to every client surface. The most direct first-party account of why approval flows, streaming diffs, and thread persistence demand a purpose-built protocol — and why MCP's tool-oriented model proved insufficient for these requirements." aiboost_harness,117,Design Primitives > Agent Loop,Hooks – Codex,https://developers.openai.com/codex/hooks,,,,,,grey,"OpenAI's lifecycle-hook framework for Codex: inject deterministic scripts at SessionStart, PreToolUse, PostToolUse, and other loop events to enforce guardrails, audit actions, and customize agent behavior without relying on prompt-level trust. A concrete reference for programmable harness governance." aiboost_harness,118,Design Primitives > Agent Loop,Extended Thinking — Claude API Docs,https://docs.anthropic.com/en/docs/build-with-claude/extended-thinking,,,,,,grey,"The harness-critical reference for integrating extended thinking into agent loops: budget_tokens controls reasoning depth per turn, thinking blocks must be preserved when passing tool results back (omitting them silently breaks multi-step reasoning), and thinking mode cannot change mid-turn. Essential before wiring extended thinking into any tool-use loop." aiboost_harness,119,Design Primitives > Agent Loop,Getting started with loops,https://claude.com/blog/getting-started-with-loops,,,2026,6,text_date,grey,"Anthropic's June 2026 practical taxonomy of agent loops: turn-based, goal-based (/goal), time-based (/loop, /schedule), and proactive loops. The framework for matching loop primitive to task shape — and the emphasis on deterministic stop conditions and token budgets — makes it a concise reference for choosing the right loop abstraction instead of defaulting to a single conversational turn cycle." aiboost_harness,120,Design Primitives > Agent Loop,Improving Deep Agents with Harness Engineering,https://blog.langchain.com/improving-deep-agents-with-harness-engineering/,,,,,,grey,"LangChain's case study showing harness-only changes moved their coding agent from rank 30 to top 5 on Terminal Bench 2.0 with no model swap: structured verification loops, context injection (directory maps + time budget warnings), loop-detection middleware, and a ""reasoning sandwich"" concentrating maximum thinking at planning and verification phases. The most concrete published demonstration that " aiboost_harness,121,Design Primitives > Agent Loop,Loop Engineering,https://github.com/cobusgreyling/loop-engineering,,,,,,repository,"Practical design system for agent loops with seven production patterns, cross-tool starter kits, and CLI tools that score readiness, scaffold state, estimate cost, detect drift, and isolate worktrees. The clearest open-source resource for moving from one-off prompting to durable, observable agent loops." aiboost_harness,122,Design Primitives > Agent Loop,Life-Harness,https://github.com/Tianshi-Xu/Life-Harness,,,,,,repository,"Official implementation of a lifecycle-aware runtime harness that improves frozen LLM agents by adapting the model-environment interface across four layers: environment contract, procedural skills, action realization, and trajectory regulation. The key result is that harness-side adaptation transfers across 18 model backbones, proving that many agent failures are interface mismatches rather than r" aiboost_harness,123,Design Primitives > Agent Loop,How Middleware Lets You Customize Your Agent Harness,https://blog.langchain.com/how-middleware-lets-you-customize-your-agent-harness/,,,,,,grey,"Introduces AgentMiddleware: six composable hooks (before_agent, before_model, wrap_model_call, wrap_tool_call, after_model, after_agent) that intercept every stage of the agent loop. Enables deterministic policy enforcement (PII redaction that can't be trusted to prompts), dynamic tool injection, mid-task model swapping, and production patterns (retry, fallback, HITL interrupts) without modifying " aiboost_harness,124,Design Primitives > Agent Loop,Agents Learn Their Runtime: Interpreter Persistence as Training-Time Semantics,https://arxiv.org/abs/2603.01209,2603.01209,,2026,3,arxiv_id,academic,Controlled experiment isolating interpreter state persistence as an independent training variable. The harness finding: mismatching your runtime persistence mode to the model's training-time semantics produces either 80% missing-variable errors (model expects state that doesn't persist) or 3.5× token overhead (model redundantly recomputes state it expects to already have). Persistence is a learned aiboost_harness,125,Design Primitives > Agent Loop,Real-Time Deadlines Reveal Temporal Awareness Failures in LLM Strategic Reasoning,https://arxiv.org/abs/2601.13206,2601.13206,,2026,1,arxiv_id,academic,"Demonstrates that temporal awareness (handling deadlines and time constraints) appears orthogonal to reasoning capability: explicit temporal feedback in the agent loop significantly improves LLM performance on deadline-constrained tasks. Indicates temporal semantics as a learned behavior that must be integrated into harness-level context (current time, deadlines, time budgets) rather than assumed " aiboost_harness,126,Design Primitives > Agent Loop,A Scheduler-Theoretic Framework for LLM Agent Execution,https://arxiv.org/abs/2604.11378,2604.11378,,2026,4,arxiv_id,academic,"April 2026 systematic analysis of 70 open-source LLM agent projects showing 60% adopt the Agent Loop pattern. Proposes a formal scheduler framework that maps execution patterns (Agent Loop, Event-driven, State-machine, Graph/flow, Hybrid) onto a unified control model, making the controllability/expressiveness/implementability trade-offs explicit. Essential reading for choosing the right loop archi" aiboost_harness,127,Design Primitives > Agent Loop,Confucius Code Agent (CCA),https://github.com/facebookresearch/cca-swebench,,,2026,2,text_date,repository,"](https://github.com/facebookresearch/cca-swebench) — February 2026 production-grade coding agent from Meta/Harvard built on the Confucius SDK, which structures harness design around three perspectives: Agent Experience (AX), User Experience (UX), and Developer Experience (DX). Features a unified orchestrator with advanced context management, persistent note-taking for cross-session learning, and " aiboost_harness,128,Design Primitives > Agent Loop,The Design Space of Today's and Future AI Agent Systems,https://arxiv.org/abs/2604.14228,2604.14228,,2026,4,arxiv_id,academic,"April 2026 reverse-engineering of Claude Code's architecture revealing five-stage progressive compaction (budget reduction → snip → microcompact → context collapse → auto-compact), subagent isolation with rebuilt permission contexts, and a 27-event-type hook pipeline. The most detailed public analysis of a production agent loop's internal design decisions — essential for understanding how context " aiboost_harness,129,Design Primitives > Agent Loop,deepclaude,https://github.com/aattaran/deepclaude,,,,,,repository,"Ports Claude Code's full agent loop to DeepSeek V4 Pro and other Anthropic-compatible backends while preserving the same UX. The strongest practical evidence that loop architecture — not model identity — determines agent behavior, and a concrete starting point for building backend-agnostic harnesses." aiboost_harness,130,Design Primitives > Agent Loop,The Coding Harness Behind GitHub Copilot in VS Code,https://code.visualstudio.com/blogs/2026/05/15/agent-harnesses-github-copilot-vscode,,,2026,5,url_date,grey,"VS Code team's breakdown of the coding harness behind GitHub Copilot: three core loop responsibilities (context assembly, tool exposure, tool execution), multi-provider model routing across Anthropic, Google, OpenAI, xAI, and Mistral, and the VSC-Bench eval suite with PR-gated assessment. The clearest published account of how a major product treats harness changes as first-class code review criter" aiboost_harness,131,Design Primitives > Agent Loop,statewright,https://github.com/statewright/statewright,,,,,,repository,"State machine guardrails that constrain which tools an agent can call in each phase of a workflow, turning open-ended loops into deterministic state transitions. The research result is striking: local models went from 2/10 to 10/10 passing on a SWE-bench subset purely by shrinking the tool space, proving that loop structure — not model size — is the binding constraint." aiboost_harness,132,Design Primitives > Agent Loop,Introducing dynamic workflows in Claude Code,https://claude.com/blog/introducing-dynamic-workflows-in-claude-code,,,2026,5,text_date,grey,"Anthropic's May 2026 introduction to dynamic parallel subagent orchestration: Claude generates JavaScript orchestration scripts that fan out work to tens or hundreds of parallel subagents with adversarial verification, converging on answers for tasks like the 750k-line Bun Zig-to-Rust port. The key harness insight is that the plan lives in executable code rather than the model's context window, sc" aiboost_harness,133,Design Primitives > Agent Loop,AgentSPEX,https://github.com/ScaleML/AgentSPEX,,,,,,repository,"UIUC's open-source specification and execution language for LLM-agent workflows: declarative YAML with typed steps, branching, loops, and explicit state management, backed by a Docker sandbox with 50+ MCP tools, checkpointing, and trajectory logging. A concrete reference for turning ad-hoc agent loops into version-controlled, reproducible harness artifacts." aiboost_harness,134,Design Primitives > Agent Loop,TomasuLLM: Out-of-Order Speculative Execution for LLM Agents,https://arxiv.org/abs/2609.38201,2609.38201,,2026,9,arxiv_id,academic,"September 2026 paper that ports speculative execution from processor design to the agent loop: the harness predicts future tool calls, runs them early in isolated copy-on-write sandboxes, and commits results in the original trajectory order only after validation — attacking the ""observation stall"" where long compiles and test suites leave the agent idle. 1.31× speedup on SWE-bench Verified and 1.3" aiboost_harness,138,Design Primitives > Planning & Task Decomposition,Run Long-Horizon Tasks with Codex,https://developers.openai.com/blog/run-long-horizon-tasks-with-codex/,,,,,,grey,"Introduces milestone-based planning artifacts (Plan.md, Implement.md) as harness-level state." aiboost_harness,139,Design Primitives > Planning & Task Decomposition,Harness Design for Long-Running Application Development,https://www.anthropic.com/engineering/harness-design-long-running-apps,,,,,,grey,"Multi-session planning, progress tracking, and the role of persistent planning documents." aiboost_harness,140,Design Primitives > Planning & Task Decomposition,Plan-and-Execute Agents,https://blog.langchain.com/plan-and-execute-agents/,,,,,,grey,"The canonical engineering write-up separating planning from execution as distinct harness layers: a planner LLM generates the step list once; an executor agent works through it, replanning only when needed. Defines the pattern that most modern task-decomposition harnesses follow." aiboost_harness,141,Design Primitives > Planning & Task Decomposition,microsoft/TaskWeaver,https://github.com/microsoft/TaskWeaver,,,,,,repository,Code-first task decomposition framework with a planner/executor split and a plugin system for injecting domain knowledge into the planning layer. The most complete reference implementation of plan-then-execute with stateful task tracking. aiboost_harness,142,Design Primitives > Planning & Task Decomposition,LATS: Language Agent Tree Search,https://arxiv.org/abs/2310.04406,2310.04406,,2023,10,arxiv_id,academic,"Unifies reasoning, acting, and planning via Monte Carlo Tree Search over agent trajectories. Directly informs harness design: external tool feedback as tree-search signals, trajectory backtracking on failure, and depth-bounded exploration make this the most actionable planning research for harnesses with real environment interaction." aiboost_harness,143,Design Primitives > Planning & Task Decomposition,Agyn: A Multi-Agent System for Team-Based Autonomous Software Engineering,https://arxiv.org/abs/2602.01465,2602.01465,,2026,2,arxiv_id,academic,"Demonstrates specialized harness patterns for coordinating heterogeneous agent teams (planner, coder, reviewer, executor) on software engineering tasks. Shows how role-specific agents with different model sizes and tool access produce better outcomes than single-agent approaches, with concrete metrics on task decomposition effectiveness." aiboost_harness,144,Design Primitives > Planning & Task Decomposition,Plan-and-Act: Improving Planning of Agents for Long-Horizon Tasks,https://arxiv.org/abs/2503.09572,2503.09572,,2025,3,arxiv_id,academic,"Modular framework separating high-level planning from low-level execution through synthetic data generation and explicit structured planning. Achieves 57.58% success on WebArena-Lite and 81.36% on WebVoyager. The key harness insight is that planner and executor can be specialized independently — different model sizes, tool access, and reasoning budgets for each layer — improving overall reliabilit" aiboost_harness,145,Design Primitives > Planning & Task Decomposition,Choosing the Right Multi-Agent Architecture,https://blog.langchain.com/choosing-the-right-multi-agent-architecture/,,,,,,grey,"Decision framework for four multi-agent patterns (subagents, skills, handoffs, router) with concrete performance data: subagents process 67% fewer tokens than skills in multi-domain scenarios because context isolation prevents cross-domain bloat. The five-dimension matching table (distributed development, parallelization, multi-hop, user interaction, latency) is the most actionable published guide" aiboost_harness,146,Design Primitives > Planning & Task Decomposition,Multi-Agent Workflows Often Fail. Here's How to Engineer Ones That Don't.,https://github.blog/ai-and-ml/generative-ai/multi-agent-workflows-often-fail-heres-how-to-engineer-ones-that-dont/,,,2026,2,text_date,grey,"GitHub's February 24, 2026 distillation of a failure pattern most harnesses eventually rediscover: multi-agent systems behave like distributed systems, so every handoff needs typed schemas, constrained action schemas, and explicit boundary validation. Worth including because it turns ""add more agents"" from a vibe into an interface design problem you can actually reason about." aiboost_harness,147,Design Primitives > Planning & Task Decomposition,Effective Harnesses for Long-Running Agents,https://www.anthropic.com/engineering/effective-harnesses-for-long-running-agents,,,,,,grey,"Anthropic's pattern for maintaining agent progress across multiple context windows: an initializer agent sets up the environment once and hands off to a coding agent that makes incremental progress each session. The structured handoff mechanism — feature lists, git commits, and test gates as cross-session state — is the reference design for any harness where a task exceeds a single context window " aiboost_harness,148,Design Primitives > Planning & Task Decomposition,Task-Adaptive Multi-Agent Orchestration (AdaptOrch),https://arxiv.org/abs/2602.16873,2602.16873,,2026,2,arxiv_id,academic,"](https://arxiv.org/abs/2602.16873) — February 2026 framework that dynamically selects orchestration topology (parallel, sequential, hierarchical, or hybrid) based on task dependency graphs rather than fixed pipeline architecture. Demonstrates that topology choice is a harness-level lever that can improve performance 12–23% over model selection alone." aiboost_harness,149,Design Primitives > Planning & Task Decomposition,GitHub Spec Kit,https://github.com/github/spec-kit,,,,,,repository,"GitHub's open-source toolkit for spec-driven development: spec.md → plan.md → tasks.md artifacts convert natural-language intent into a machine-checkable plan before any code is written, with /specify, /plan, /tasks, and /implement commands that run across Claude Code, Copilot, Codex, and Gemini CLI. The clearest reference implementation of treating spec/plan/task documents — not the prompt — as t" aiboost_harness,150,Design Primitives > Planning & Task Decomposition,Task-Decoupled Planning for Long-Horizon Agents (TDP),https://arxiv.org/abs/2601.07577,2601.07577,,2026,1,arxiv_id,academic,"](https://arxiv.org/abs/2601.07577) — January 2026 planning framework that combines task decomposition with modular agent design: a Supervisor decomposes tasks into a dependency graph, Planner & Executor agents solve each decoupled sub-task node independently, and a Self-Revision module updates the graph after execution. The key harness insight is that decoupling planning from execution at the sub" aiboost_harness,151,Design Primitives > Planning & Task Decomposition,"Fresh Memory, Stale Plans: Derivation Currency for Distributed LLM-Agent Memory",https://arxiv.org/abs/2609.03340,2609.03340,,2026,9,arxiv_id,academic,"September 2026 paper naming the failure mode every freshness check misses: an agent can inherit the *latest* requirement through shared memory while executing a plan derived from inputs that have since moved on. Timestamp/head comparisons check *observation* currency (is my copy current?) rather than *derivation* currency (are the inputs this plan was computed from still current?), which is why su" aiboost_harness,155,Design Primitives > Context Delivery & Compaction,Harness Engineering,https://openai.com/index/harness-engineering/,,,,,,grey,"How to structure context windows for agents: what to include, what to exclude, and how context shape affects agent behavior." aiboost_harness,156,Design Primitives > Context Delivery & Compaction,Effective Context Engineering for AI Agents,https://www.anthropic.com/engineering/effective-context-engineering-for-ai-agents,,,,,,grey,"Anthropic's systematic guide to managing the full context state—system prompts, tools, MCP, and message history—as a finite, curated resource. Reframes harness design as ""what configuration of context produces the desired behavior?"" rather than just prompt wording." aiboost_harness,157,Design Primitives > Context Delivery & Compaction,Compaction — Claude API Docs,https://platform.claude.com/docs/en/build-with-claude/compaction,,,,,,grey,Anthropic's reference for server-side context compaction: automatically summarizes older context when approaching the window limit. Reduced token consumption by 84% in a 100-turn web search eval while allowing agents to complete workflows that would otherwise hit context limits. aiboost_harness,158,Design Primitives > Context Delivery & Compaction,LLMLingua,https://github.com/microsoft/LLMLingua,,,,,,repository,"Microsoft Research's prompt compression toolkit (up to 20x compression, minimal performance loss) that can be embedded as a preprocessing step in the context delivery layer. LLMLingua-2 adds 3–6x speed gains, making it viable for latency-sensitive agent loops." aiboost_harness,159,Design Primitives > Context Delivery & Compaction,Prompt Caching — Claude API Docs,https://docs.anthropic.com/en/docs/build-with-claude/prompt-caching,,,,,,grey,"The most effective harness-level cost lever: cache repeated system prompts, tool definitions, and long documents across requests. Explains where to place cache_control breakpoints for maximum reuse across multi-turn agent sessions." aiboost_harness,160,Design Primitives > Context Delivery & Compaction,Autonomous Context Compression,https://blog.langchain.com/autonomous-context-compression/,,,,,,grey,Shifts context compression from harness-controlled (compacting at a fixed token threshold) to agent-controlled: agents call a dedicated tool to trigger compression when strategically appropriate — between tasks or before consuming large inputs. Eliminates the failure mode where reactive-at-limit compaction interrupts agents mid-subtask and corrupts in-flight reasoning state. aiboost_harness,161,Design Primitives > Context Delivery & Compaction,Active Context Compression: Autonomous Memory Management in LLM Agents,https://arxiv.org/abs/2601.07190,2601.07190,,2026,1,arxiv_id,academic,"Proposes a ""Focus Agent"" architecture where the agent autonomously decides when to consolidate interaction history into a persistent Knowledge block and prune raw context — shifting compression from a harness-enforced policy to a model-controlled action. Produces 22.7% token reduction with no accuracy loss on long-horizon tasks; the core contribution is making the compression unit semantically coh" aiboost_harness,162,Design Primitives > Context Delivery & Compaction,context-mode,https://github.com/mksglu/context-mode,,,,,,repository,"MCP server that intercepts raw tool output before it enters the context window, sandboxing bulky data (Playwright snapshots, GitHub issues, logs) outside the LLM and retrieving only relevant fragments via BM25 when needed. The ""think in code"" paradigm — replacing ten file-read tool calls with one script execution — is a concrete harness pattern for turning context pressure into a programming probl" aiboost_harness,163,Design Primitives > Context Delivery & Compaction,Making Agent-Friendly Pages with Content Negotiation,https://vercel.com/blog/making-agent-friendly-pages-with-content-negotiation,,,2026,2,text_date,grey,"Vercel's February 3, 2026 implementation guide for serving text/markdown when agents request it via Accept: text/markdown, while preserving the same human-facing HTML URL. This is a real harness primitive, not just a docs trick: it removes boilerplate before it ever enters the context window and gives agents cleaner, cheaper inputs without custom scrapers." aiboost_harness,164,Design Primitives > Context Delivery & Compaction,A-RAG: Scaling Agentic Retrieval-Augmented Generation via Hierarchical Retrieval Interfaces,https://arxiv.org/abs/2602.03442,2602.03442,,2026,2,arxiv_id,academic,"Reframes RAG as a harness tool-design problem: instead of injecting retrieved documents into context at pipeline time, expose three retrieval tools (keyword search, semantic search, chunk read) and let the agent pull information incrementally as each reasoning step requires it. The key harness decision is architectural — retrieval becomes a tool call in the agent loop, not a preprocessing step — w" aiboost_harness,165,Design Primitives > Context Delivery & Compaction,"LLM Readiness Harness: Evaluation, Observability, and CI Gates for LLM/RAG Applications",https://arxiv.org/abs/2603.27355,2603.27355,,2026,3,arxiv_id,academic,"Structured framework for building production-grade evaluation harnesses: evaluation gates that block deployment, observability instrumentation that tracks all agent decisions, and CI integration patterns that catch regressions before they reach users. Essential reading for organizations deploying multiple agents in parallel where a single harness failure can cascade." aiboost_harness,166,Design Primitives > Context Delivery & Compaction,ByteRover: Agent-Native Memory Through LLM-Curated Hierarchical Context,https://arxiv.org/abs/2604.01599,2604.01599,,2026,4,arxiv_id,academic,LLM-curated hierarchical context management for agents where the model itself learns to weight information importance across multiple hierarchy levels. Reduces token overhead through learned relevance filtering without sacrificing comprehension. Directly applicable to any harness where context budget is the limiting factor — letting the model curate what belongs in active memory vs. what can be re aiboost_harness,167,Design Primitives > Context Delivery & Compaction,Claude Code Compaction: How Context Compression Works,https://okhlopkov.com/claude-code-compaction-explained/,,,2026,3,text_date,grey,"March 2026 deep-dive into Claude Code's automatic compaction mechanism: what survives (current task, recent errors, file names) vs. what gets lost (initial instructions, intermediate decisions, style rules). Key harness insight: never rely on compaction for critical rules — move them to CLAUDE.md where they live in the system prompt and survive any compression. Essential practical guidance for any" aiboost_harness,168,Design Primitives > Context Delivery & Compaction,Token Savior,https://github.com/Mibayy/token-savior,,,,,,repository,"MCP server that indexes codebases by symbol (functions, classes, call graphs) so agents navigate by pointer instead of reading whole files, cutting active tokens by 77% and benchmark wall time by 76%. Demonstrates that context delivery for coding agents is a navigation problem, not just a compression problem." aiboost_harness,169,Design Primitives > Context Delivery & Compaction,Trellis,https://github.com/mindfold-ai/Trellis,,,,,,repository,"Replaces the bloated CLAUDE.md pattern with a progressive spec system: agents load only the standards, task PRDs, and session journals relevant to the current step. The cross-platform adapter layer turns vendor-specific harness configuration into a portable team practice rather than a per-tool hack." aiboost_harness,170,Design Primitives > Context Delivery & Compaction,OpenViking,https://github.com/volcengine/OpenViking,,,,,,repository,"ByteDance/Volcengine's open-source context database that puts an agent's knowledge, memory, and skills in one virtual filesystem under viking://, navigated with ls/tree/read/grep and inspectable by humans instead of hidden in an embedding store. Every directory carries generated L0 abstracts and L1 overviews so agents read summaries before opening full content (L2), retrieval is scoped to a direct" aiboost_harness,171,Design Primitives > Context Delivery & Compaction,DESIGN.md,https://github.com/google-labs-code/design.md,,,,,,repository,"Google Labs' specification for describing visual identity systems to coding agents: machine-readable design tokens (YAML front matter) combined with human-readable design rationale (markdown prose) give agents a persistent, structured understanding of design constraints without requiring custom tool chains." aiboost_harness,172,Design Primitives > Context Delivery & Compaction,codebase-memory-mcp,https://github.com/DeusData/codebase-memory-mcp,,,,,,repository,"High-performance code intelligence MCP server that full-indexes repositories into a persistent knowledge graph via tree-sitter AST analysis across 66 languages. Replaces dozens of file-read/grep cycles with sub-millisecond structured queries, cutting active tokens by 120× and turning codebase navigation from a context-pressure problem into a pointer-chasing problem." aiboost_harness,173,Design Primitives > Context Delivery & Compaction,Mirage,https://github.com/strukto-ai/mirage,,,,,,repository,"Mounts S3, Slack, Gmail, GitHub, and Redis side-by-side as a single virtual filesystem so agents interact with every backend through familiar bash commands instead of learning N distinct APIs. The key harness insight: LLMs are already fluent in grep, cat, and cp — leveraging that vocabulary eliminates tool-schema bloat and makes cross-service pipelines compose as naturally as local shell scripts." aiboost_harness,174,Design Primitives > Context Delivery & Compaction,dirac,https://github.com/dirac-run/dirac,,,,,,repository,"Coding agent harness optimized for surgical context curation and API cost reduction: Hash Anchored edits, massively parallel operations, and AST manipulation combine to cut costs 50–80% while improving code quality. Demonstrates that precise context delivery — not just bulk compression — is the right lever for efficient coding agents." aiboost_harness,175,Design Primitives > Context Delivery & Compaction,MinishLab/semble,https://github.com/MinishLab/semble,,,,,,repository,"Code search primitive that replaces grep+read cycles with natural-language retrieval, cutting active tokens by ~98% while keeping 99% of a transformer-based retriever's accuracy. Ships as an MCP server and CLI, runs on CPU with zero external dependencies — the right drop-in for any coding agent harness struggling with context pressure." aiboost_harness,176,Design Primitives > Context Delivery & Compaction,harness-experimental,https://github.com/hoangnb24/harness-experimental,,,,,,repository,"Repository-level operating harness that turns any software repo into an agent-ready workspace: structured AGENTS.md, HARNESS.md, and FEATURE_INTAKE.md give agents the missing project context — where to start, what the product contract says, how risky the change is, and which decisions future agents should inherit. The most concrete open-source implementation of ""coding agents need better repositor" aiboost_harness,177,Design Primitives > Context Delivery & Compaction,headroom,https://github.com/chopratejas/headroom,,,,,,repository,"Compresses tool outputs, logs, files, and RAG chunks before they enter the context window, cutting active tokens by 60–95% without changing answers. Ships as a library, proxy, and MCP server — the right drop-in layer for any harness where bulky tool returns are the primary context pressure source." aiboost_harness,178,Design Primitives > Context Delivery & Compaction,Context7,https://github.com/upstash/context7,,,,,,repository,"MCP server and CLI that injects up-to-date, version-specific library documentation directly into agent context, eliminating hallucinated APIs and outdated code examples caused by stale training data. Ships as both a ctx7 command-line tool and an MCP server with resolve-library-id and query-docs tools." aiboost_harness,179,Design Primitives > Context Delivery & Compaction,Context Pruning for Coding Agents via Multi-Rubric Latent Reasoning,https://arxiv.org/abs/2605.15315,2605.15315,,2026,5,arxiv_id,academic,"Decomposes code relevance into two interpretable dimensions — semantic evidence and dependency support — rather than collapsing all retention decisions into a single score. Saves up to 31% more tokens on multi-turn coding agent tasks while improving Exact Match by up to +3.5, demonstrating that coding-agent context pruning needs domain-specific rubrics rather than generic compression." aiboost_harness,180,Design Primitives > Context Delivery & Compaction,OpenWiki,https://github.com/langchain-ai/openwiki,,,,,,repository,"LangChain's CLI that writes and maintains agent-readable wikis for codebases or purpose memory, turning documentation drift into a versioned, automatable harness artifact. Emits Google Open Knowledge Format bundles so curated context stays portable across agents and can be kept fresh via CI." aiboost_harness,181,Design Primitives > Context Delivery & Compaction,PRO-LONG,https://github.com/alexisfox7/PRO-LONG,,,,,,repository,"Programmatic memory framework for long-horizon agents: the harness appends all observations to a structured log and lets the agent search it with code instead of relying on fixed summarization or compaction policies. Achieves 4.2–5.8× token reduction and matches or exceeds specialized harnesses on ARC-AGI-3, showing that long-horizon context management can be a programmable retrieval problem rathe" aiboost_harness,182,Design Primitives > Context Delivery & Compaction,Graft,https://github.com/NanoNets/Graft,,,,,,repository,"Builds a local, regenerable graph of plain-English system explanations and code relationships, then rides along inside Claude Code, Cursor, Codex, and Gemini via MCP and statusline hooks so the agent stops rediscovering the repo every session. The published SWE-bench Verified and efficiency benchmarks make it the clearest recent demonstration that context delivery for coding agents is a navigation" aiboost_harness,184,Design Primitives > Context Delivery & Compaction,Organizing Context in a Multi-Agent Harness,https://www.langchain.com/blog/organizing-context-in-a-multi-agent-harness,,,2026,9,text_date,grey,"LangChain's September 2026 introduction of context modes for subagents: isolated subagents start fresh (context isolation as a firewall), while forked subagents inherit the supervisor's full conversation — excised trailing tool call, task description rewritten as a user message, prompt caching preserved by design. The worker-vs-verifier decision rule — fork workers that continue in-progress invest" aiboost_harness,185,Design Primitives > Context Delivery & Compaction,SoL-Pi,https://github.com/NVlabs/SoL-Pi,,,2026,9,text_date,repository,"NVIDIA's September 2026 extension for the Pi harness that packages four efficiency mechanisms discovered by running 152 candidate ideas through auto-research loops: Action Fusion (an edit runs its follow-up validation in the same tool call), ObservationPack (repeated large tool results become stable handles with exact paged recall), an Evidence-Preserving Reducer (long logs compact to receipts onl" aiboost_harness,186,Design Primitives > Context Delivery & Compaction,"Boris Cherny: Building Claude Code (YC Startup School, July 2026)",https://www.ycombinator.com/library/UN-boris-cherny-building-claude-code,,,2026,7,text_date,grey,"](https://www.ycombinator.com/library/UN-boris-cherny-building-claude-code) — The Claude Code creator's account of shipping Opus 5: Anthropic deleted ~80% of the system prompt because it encoded corrections for behaviors the model no longer has, with no eval regression. The ablation methodology — strip the prompt entirely, add back only lines that fix repeated failures, repeat at every model relea" aiboost_harness,187,Design Primitives > Context Delivery & Compaction,"Context management in agent harnesses: memory, files, and subagents",https://arize.com/blog/context-management-in-agent-harnesses/,,,2026,4,text_date,grey,"Arize's April 2026 comparative teardown of how Pi, OpenClaw, Claude Code, and Letta each manage context as an active resource: hard read caps with pagination, LLM compaction triggered at token thresholds with tool-call boundary safety, oversized-result persistence to disk, and subagent context isolation. The striking finding is convergence — four independent harnesses plus Arize's own production a" aiboost_harness,188,Design Primitives > Context Delivery & Compaction,Context Engineering in 2026: Why We Stopped Compacting Our Agent's Context,https://www.louisbouchard.ai/context-engineering-2026/,,,2026,8,text_date,grey,"August 2026 controlled study on a production AI tutor (660 turns, planted-fact probes) that inverts the compaction default: keeping full history reached 100% memory recall at $0.11/turn while the compacting setup managed 58% recall at $0.24/turn, because summarization rewrites the cached prefix and destroys cache hits. The decision framework — name the real constraint first (window size → retrieva" aiboost_harness,189,Design Primitives > Context Delivery & Compaction,AutoCompact: Learning When to Compact Context in Long-Horizon Coding Agents,https://arxiv.org/abs/2610.02163,2610.02163,,2026,10,arxiv_id,academic,"October 2026 paper that turns compaction timing from a fixed harness policy into a learned agent behavior: a judge corrects compaction timing, summary quality, and post-compaction actions in collected trajectories, then supervised fine-tuning plus joint RL optimizes coding and compaction together. +9.2% absolute on SWE-bench Verified and +5.0% on SWE-PolyBench Verified — including in a 256K window" aiboost_harness,190,Design Primitives > Context Delivery & Compaction,The Compaction Cliff in Long-Running AI Agent Memory,https://arxiv.org/abs/2608.22752,2608.22752,,2026,8,arxiv_id,academic,"CIKM 2026 paper measuring what compaction actually destroys: Claude Code's /compact prompt on Sonnet 4.6 retains only 53% of safety rules after one round and 10% after five, with the same decay across four model families and every summarization strategy tested. The proposed Knowledge Triage classifies each knowledge line by type (constraint, procedural, belief, preference, episodic) and routes eac" aiboost_harness,191,Design Primitives > Context Delivery & Compaction,Diagnosing and Mitigating Context Rot in Long-horizon Search,https://arxiv.org/abs/2606.29718,2606.29718,,2026,6,arxiv_id,academic,"GAIR/SJTU paper that names the failure mode compaction debates usually miss: *premature termination*, where search agents give up or answer uncertainly on problems they can solve, with the rate rising as context accumulates — long before the window fills. Pruning experiments show content matters more than length (dropping history removes give-ups but causes no-answer spikes, so noise is the toxin," aiboost_harness,195,Design Primitives > Tool Design,"A failed network check is a finding, not a tool crash",https://gist.github.com/seifip/fa05e77888fc8ec0e99463084f056ae6#file-readme-md,,,,,,repository,"A worked Python adapter separates a diagnostic finding from an execution error when a CLI returns valid JSON with exit code 1. It shows how to bind target configuration outside model arguments, preserve skipped checks, and minimize returned fields, with local test results and explicit limits." aiboost_harness,196,Design Primitives > Tool Design,Writing Effective Tools for Agents,https://www.anthropic.com/engineering/writing-tools-for-agents,,,,,,grey,"Tool naming, schema design, error messages, and return value conventions that make agents more reliable." aiboost_harness,197,Design Primitives > Tool Design,Tool Use — Claude API Docs,https://platform.claude.com/docs/en/agents-and-tools/tool-use/overview,,,,,,grey,"Authoritative reference for client vs. server tool execution models, strict schema enforcement, and tool_result error signaling. The distinction between client-side and server-side tool execution is a foundational harness architecture decision." aiboost_harness,198,Design Primitives > Tool Design,Function Calling — OpenAI Docs,https://platform.openai.com/docs/guides/function-calling,,,,,,grey,Defines the de facto industry-standard JSON Schema conventions for tool definitions and parallel function calling. Essential reading before designing a tool interface that needs to work across multiple models. aiboost_harness,199,Design Primitives > Tool Design,Tool Annotations as Risk Vocabulary,https://blog.modelcontextprotocol.io/posts/2026-03-16-tool-annotations/,,,2026,3,url_date,grey,"The MCP team's definitive post on the four tool annotation hints (readOnlyHint, destructiveHint, idempotentHint, openWorldHint) as inputs to harness permission decisions, not enforced contracts. The ""lethal trifecta"" — private data access + untrusted content exposure + external communication — is the most actionable framing for why single-tool safety analysis misses the risk that emerges from tool" aiboost_harness,200,Design Primitives > Tool Design,outlines,https://github.com/dottxt-ai/outlines,,,,,,repository,"Constrains token sampling via regex/CFG/JSON Schema at the decoding layer, guaranteeing structured output without model fine-tuning. The right solution when you need OpenAI Structured Outputs-equivalent reliability from a locally deployed or open-weight model." aiboost_harness,201,Design Primitives > Tool Design,instructor,https://python.useinstructor.com/,,,,,,grey,"Maps Pydantic models directly to structured LLM extraction with built-in retry and validation-error feedback loops. Turns tool call output parsing from ad-hoc JSON handling into type-safe data models, eliminating an entire class of harness parsing bugs." aiboost_harness,202,Design Primitives > Tool Design,SkillTester: Benchmarking Utility and Security of Agent Skills,https://arxiv.org/abs/2603.28815,2603.28815,,2026,3,arxiv_id,academic,"Framework for evaluating agent skills on three dimensions (capability, robustness, security) before deployment. Directly addresses the harness problem of skill sprawl: as agents gain access to more tools, the combinatorial explosion of failure modes becomes unmanageable without systematic verification. The 86-task benchmark across 11 domains provides reference metrics for skill quality." aiboost_harness,203,Design Primitives > Tool Design,AutoHarness: Improving LLM Agents by Automatically Synthesizing a Code Harness,https://arxiv.org/abs/2603.03329,2603.03329,,2026,3,arxiv_id,academic,Google DeepMind technique that uses code synthesis to auto-generate runtime constraint harnesses from tool schemas and task specifications. Gemini-2.5-Flash + AutoHarness outperforms Gemini-2.5-Pro and GPT-5.2-High on TextArena games by eliminating illegal moves through learned harness policies. Shifts constraint enforcement from static (schema validation) to dynamic (synthesized code guards) — a aiboost_harness,204,Design Primitives > Tool Design,Scaling Parallel Tool Calling for Efficient Deep Research,https://arxiv.org/abs/2602.07359,2602.07359,,2026,2,arxiv_id,academic,"February 2026 analysis of how parallel tool calling reduces latency in multi-step agent workflows. Demonstrates that concurrent tool execution (rather than sequential observe→act loops) is the key efficiency lever for deep-research harnesses where each step may invoke search, browse, and compute tools simultaneously. Essential for designing low-latency agent loops without sacrificing reasoning dep" aiboost_harness,205,Design Primitives > Tool Design,EigentSearch-Q+,https://arxiv.org/abs/2604.07927,2604.07927,,2026,4,arxiv_id,academic,"April 2026 framework for deep-research agents using dedicated reasoning tools (plan_next_searches, select_query_and_search, extract_relevant_details, analyze_search_progress) that externalize intermediate decisions as typed tool arguments. Inspired by Anthropic's think-tool paradigm, Q+ makes cognitive scaffolding explicit and auditable — bridging classic information-retrieval strategies with stru" aiboost_harness,206,Design Primitives > Tool Design,TopoCurate: Modeling Interaction Topology for Tool-Use Agent Training,https://arxiv.org/abs/2603.01714,2603.01714,,2026,3,arxiv_id,academic,"March 2026 framework that models interaction topology — the structural patterns of how agents invoke, chain, and conditionally branch between tools — as a first-class training signal. Rather than treating tool use as isolated function calls, TopoCurate learns topological priors from expert trajectories, improving generalization to novel tool combinations and multi-step orchestration patterns. Dire" aiboost_harness,207,Design Primitives > Tool Design,Design Patterns for Deploying AI Agents with Model Context Protocol,https://arxiv.org/abs/2603.13417,2603.13417,,2026,3,arxiv_id,academic,"March 2026 field report from an enterprise MCP deployment identifying three protocol-level gaps that break production: missing identity propagation (who is the request for?), absent adaptive tool budgeting, and unstructured error semantics. The concrete mitigation patterns — JWT-enriched tool calls, per-tool timeout contracts, and standardized error-action mappings — are essential before betting o" aiboost_harness,208,Design Primitives > Tool Design,tui-use,https://github.com/onesuper/tui-use,,,,,,repository,"Expands the agent tool surface beyond non-interactive commands: programmable TUI interaction for REPLs, debuggers, and ncurses apps that standard bash can't reach. A concrete harness primitive for any agent that needs to operate interactive CLI tools without building a custom wrapper per program." aiboost_harness,209,Design Primitives > Tool Design,CLI-Anything,https://github.com/HKUDS/CLI-Anything,,,,,,repository,"Generates agent-native CLI harnesses for any software, giving agents structured JSON access to applications that were never designed for automation. The CLI-Hub registry and auto-generated SKILL.md files turn tool expansion into a package-manager experience — solving the ""long tail"" of agent tool coverage without a custom wrapper per program." aiboost_harness,210,Design Primitives > Tool Design,zerolang,https://github.com/vercel-labs/zerolang,,,,,,repository,"Experimental graph-first programming language where agents inspect and edit code through a compiler-derived ProgramGraph (node IDs, graph hashes, types, effects, ownership) instead of fragile text patches. Collapses the typical agent loop of edit-format-reparse-check-fix into a single compiler-validated semantic operation — a reference design for making code manipulation a structured tool interfac" aiboost_harness,211,Design Primitives > Tool Design,The Devil Is in the Interface: Evaluating How Tool Architecture Shapes Coding Agent Behavior,https://arxiv.org/abs/2608.11386,2608.11386,,2026,8,arxiv_id,academic,"August 2026 controlled study (11,700 trajectories, six tool architectures with capabilities held constant) isolating tool *architecture* — how capabilities are organized and exposed — as an independent variable. The findings give tool design its first rigorous dose-response evidence: structured low-level interfaces improve consistency up to 4.7×, natural-language search broadens repo exploration 1" aiboost_harness,217,Design Primitives > Skills & MCP,Model Context Protocol,https://modelcontextprotocol.io/introduction,,,,,,grey,"Anthropic's open protocol for connecting agents to external tools, data sources, and services in a standardized way." aiboost_harness,218,Design Primitives > Skills & MCP,modelcontextprotocol/servers,https://github.com/modelcontextprotocol/servers,,,,,,repository,"Anthropic's official reference MCP server implementations (GitHub, Slack, Postgres, Puppeteer, etc.). The authoritative source for understanding correct MCP server structure before building your own." aiboost_harness,219,Design Primitives > Skills & MCP,microsoft/playwright-mcp,https://github.com/microsoft/playwright-mcp,,,,,,repository,"Browser automation via accessibility tree snapshots rather than screenshots, dramatically reducing token cost. The canonical example of structured tool output design in an MCP server." aiboost_harness,220,Design Primitives > Skills & MCP,Chrome DevTools MCP,https://github.com/ChromeDevTools/chrome-devtools-mcp,,,,,,repository,"Official Google MCP server that exposes live Chrome debugging surfaces — network analysis, performance profiling, console messages, memory snapshots, and Lighthouse audits — as structured agent tools. The clearest reference for turning browser inspection into a first-class tool interface rather than relying solely on screenshot-driven automation." aiboost_harness,221,Design Primitives > Skills & MCP,agent-device,https://github.com/callstackincubator/agent-device,,,,,,repository,"MCP-native control layer for iOS and Android devices: snapshots, semantic targeting, typed client access, diagnostics, and replayable workflows. Fills a critical gap in the mobile-agent harness stack — most tool design assumes desktop or browser surfaces, but real-world agents increasingly need to interact with native mobile apps." aiboost_harness,222,Design Primitives > Skills & MCP,A2A Protocol,https://github.com/a2aproject/A2A,,,,,,repository,Google's open Agent-to-Agent protocol: JSON-RPC over HTTP(S)/SSE with Agent Card service discovery and a task/message/artifact communication model. The emerging standard for cross-framework agent interoperability in multi-agent harnesses. aiboost_harness,223,Design Primitives > Skills & MCP,Announcing the Agentic Resource Discovery specification,https://developers.googleblog.com/announcing-the-agentic-resource-discovery-specification/,,,2026,6,text_date,grey,"Google's June 2026 open specification for publishing, discovering, and verifying AI capabilities across the web via domain-owned catalogs and searchable registries. Adds the missing discovery layer that lets agents find MCP servers, A2A agents, and OpenAPI tools at runtime rather than relying on hardcoded integrations, with trust manifests and namespaced URNs for governance." aiboost_harness,224,Design Primitives > Skills & MCP,MCP Inspector,https://github.com/modelcontextprotocol/inspector,,,,,,repository,"Interactive debugging UI for MCP servers: inspect tool definitions, send test calls, and validate responses without wiring up a full agent. The essential development tool for anyone building or integrating MCP servers into a harness." aiboost_harness,225,Design Primitives > Skills & MCP,Shell + Skills + Compaction: Tips for Long-Running Agents,https://developers.openai.com/blog/skills-shell-tips,,,,,,grey,"OpenAI's engineering guide to three production harness primitives: versioned Skill bundles (SKILL.md manifest; routing accuracy improved 73%→85% by adding negative examples), a managed shell container for durable tool execution, and server-side compaction via explicit /responses/compact endpoint. The most concrete first-party documentation of skills-based routing and compaction published in 2026." aiboost_harness,226,Design Primitives > Skills & MCP,Composio,https://github.com/ComposioHQ/composio,,,,,,repository,"Wraps 250+ SaaS APIs (GitHub, Slack, Linear, Notion, etc.) as agent-ready actions with managed OAuth, so tool integration becomes a one-line import rather than a custom harness component per service. The fastest path from ""the agent needs to call an external API"" to a production-grade, authenticated tool." aiboost_harness,227,Design Primitives > Skills & MCP,oomol-lab/open-connector,https://github.com/oomol-lab/open-connector,,,,,,repository,"Open-source auth gateway connecting 1000+ SaaS providers to AI agents through SDK, CLI, MCP, HTTP, and OpenAPI. Treats external-tool onboarding as a unified, self-hosted harness layer: one credential and access model covers direct SDK calls, MCP servers, and OpenAPI discovery, so teams don't rebuild auth per integration." aiboost_harness,228,Design Primitives > Skills & MCP,MCP Streamable HTTP Transport,https://modelcontextprotocol.io/specification/2025-11-25/basic/transports,,,2025,11,url_date,grey,"The transport that replaced HTTP+SSE in the 2025-11-25 spec, enabling MCP servers to run as remote services rather than local processes. Servers handle multiple client connections using HTTP POST (for client→server messages) and optional GET (for server→client SSE streams). The key harness architecture decision: Streamable HTTP unlocks remote MCP deployment but introduces session management comple" aiboost_harness,229,Design Primitives > Skills & MCP,The 2026 MCP Roadmap,https://blog.modelcontextprotocol.io/posts/2026-mcp-roadmap/,,,,,,grey,"The MCP team's roadmap for the next spec cycle: horizontal-scaling transport without stateful session constraints, .well-known discovery for capability advertisement without live connections, Tasks primitive with retry/expiry semantics, and enterprise extensions (audit trails, SSO, gateway behavior). Essential reading before investing heavily in MCP server infrastructure — the transport and discov" aiboost_harness,230,Design Primitives > Skills & MCP,The 2026-07-28 MCP Specification Release Candidate,https://blog.modelcontextprotocol.io/posts/2026-07-28-release-candidate/,,,2026,7,url_date,grey,"The largest revision of MCP since launch: a stateless protocol core drops the initialize handshake and Mcp-Session-Id, the new ext-* extension framework formalizes Tasks and MCP Apps, and a twelve-month deprecation policy gives harness builders a stable target. Essential reading before designing remote MCP server infrastructure that must survive load balancers and horizontal scaling." aiboost_harness,231,Design Primitives > Skills & MCP,Developer's Guide to AI Agent Protocols,https://developers.googleblog.com/en/developers-guide-to-ai-agent-protocols/,,,,,,grey,"Google's survey of six standardized agent interoperability protocols, each solving a distinct harness integration problem: MCP (tool/data connectivity), A2A (inter-agent routing via Agent Card discovery at well-known URLs), UCP (commerce workflows), AP2 (payment authorization with spend limits), A2UI (agent-driven dynamic UIs), AG-UI (streaming event format). The most practical map of which protoc" aiboost_harness,232,Design Primitives > Skills & MCP,AG-UI,https://github.com/ag-ui-protocol/ag-ui,,,,,,repository,"Lightweight event-driven protocol standardizing how AI agents connect to frontend applications: streaming state updates, tool call rendering, and HITL interrupts over a shared event bus. Fills the layer between MCP (tool access) and A2A (agent-to-agent) — it's the missing protocol for real-time agent-to-UI communication that neither MCP nor A2A was designed to address." aiboost_harness,233,Design Primitives > Skills & MCP,Agent Client Protocol (ACP),https://github.com/agentclientprotocol/agent-client-protocol,,,,,,repository,"](https://github.com/agentclientprotocol/agent-client-protocol) — Zed and JetBrains' open standard (Apache-2.0) applying the LSP model to coding agents: JSON-RPC 2.0 over stdio, where the editor launches the agent as a subprocess and drives sessions, streaming updates, file diffs, and permission requests through one interface. With 60+ registered agents and 12+ editor clients by late 2026, it solv" aiboost_harness,234,Design Primitives > Skills & MCP,Code Execution with MCP: Building More Efficient Agents,https://www.anthropic.com/engineering/code-execution-with-mcp,,,,,,grey,Anthropic's engineering account of reducing tool-call token overhead by having agents write code to interact with MCP servers rather than calling tools directly: up to 98.7% token reduction in experiments. Broadly applicable to any harness where tool schema overhead and intermediate results are consuming context — the pattern is to wrap multi-step tool interaction in a code execution primitive rat aiboost_harness,235,Design Primitives > Skills & MCP,Microsoft Skills Framework,https://github.com/microsoft/skills,,,,,,repository,"Standardized framework for defining, versioning, and distributing agent skills. Enables skill reuse across Claude Code, Copilot, VS Code, Gemini, and other platforms — a harness-level abstraction that makes skills first-class deployment artifacts rather than ad-hoc tool definitions." aiboost_harness,236,Design Primitives > Skills & MCP,SkillNet & SkillsBench: Infrastructure for AI Agent Skills at Scale,https://github.com/skillmatic-ai/awesome-agent-skills,,,,,,repository,"Comprehensive framework for creating, evaluating, and sharing agent skills with 86-task benchmark across 11 domains. Demonstrates the harness problem of skill fragmentation and provides infrastructure for standardized skill evaluation across frameworks." aiboost_harness,237,Design Primitives > Skills & MCP,Comet,https://github.com/rpamis/comet,,,,,,repository,"Resumable long-running task workflow and Skill platform for coding that turns skill creation, evaluation, and release into a single lifecycle with Rubric, Pass@k, and Pass^k scoring. The Native/Classic dual-workflow model is a concrete reference for matching harness constraint strength to model capability rather than using one loop for every task." aiboost_harness,238,Design Primitives > Skills & MCP,addyosmani/agent-skills,https://github.com/addyosmani/agent-skills,,,,,,repository,"Production-grade engineering skills for AI coding agents, packaged as 24 reusable skills covering the full development lifecycle from /spec to /ship. The slash-command interface and context-aware auto-activation make it a concrete reference for turning senior-engineering judgment into agent-executable harness artifacts." aiboost_harness,239,Design Primitives > Skills & MCP,AWS Bedrock AgentCore with WebRTC Support,https://aws.amazon.com/about-aws/whats-new/2026/03/amazon-bedrock-webrtc/,,,2026,3,url_date,grey,"Adds peer-to-peer, UDP-based WebRTC bidirectional streaming to Bedrock Agents for real-time voice interactions. Complements existing WebSocket support with lower latency and better resilience for poor network conditions. Essential harness-level transport choice for agents targeting sub-800ms Total Turn-Around Time voice interactions." aiboost_harness,240,Design Primitives > Skills & MCP,Hermes Agent: Unified Streaming for Real-Time Agent Workflows,https://juliangoldie.com/hermes-agent-unified-streaming/,,,,,,grey,"Token-by-token streaming delivery system enabling real-time agent responses; sub-second decision loops on streaming events vs. batch-refreshed data. Critical infrastructure for harnesses where latency (not just throughput) is the constraint — agents must react to events as they arrive, not wait for batch completions." aiboost_harness,241,Design Primitives > Skills & MCP,Google Developers: Closing the Knowledge Gap with Agent Skills,https://developers.googleblog.com/closing-the-knowledge-gap-with-agent-skills/,,,,,,grey,"Google ADK expansion with evaluation harness (117 prompts) for assessing skill performance across agentic coding, chatbots, document processing. Provides reference patterns and benchmark datasets for skill evaluation, complementing the Microsoft Skills Framework with Google's evaluation methodology." aiboost_harness,242,Design Primitives > Skills & MCP,What's New with GitHub Copilot Coding Agent,https://github.blog/ai-and-ml/github-copilot/whats-new-with-github-copilot-coding-agent/,,,2026,2,text_date,grey,"GitHub's February 26, 2026 update is worth including for one specific reason: it makes .github/agents/ custom agent files, self-review, built-in security scanning, and CLI handoff concrete as harness primitives rather than abstract ideas. Useful as a current reference for how repository-scoped agent definitions and security checks are being productized in a real coding-agent control plane." aiboost_harness,243,Design Primitives > Skills & MCP,Announcing Official MCP Support for Google Services,https://cloud.google.com/blog/products/ai-machine-learning/announcing-official-mcp-support-for-google-services,,,,,,grey,"Google's 2026 rollout of managed MCP endpoints is a useful counterpoint to self-hosted MCP servers: discovery, IAM, audit logging, and Model Armor are provided as platform primitives instead of being rebuilt per server. Worth including because it shows what ""enterprise MCP"" looks like when the transport, auth, and governance layers are treated as product surface rather than glue code." aiboost_harness,244,Design Primitives > Skills & MCP,Dataverse Skills: Your Coding Agent Now Speaks Dataverse,https://devblogs.microsoft.com/powerplatform/dataverse-skills-your-coding-agent-now-speaks-dataverse,,,2026,4,text_date,grey,"Microsoft's April 1, 2026 release is a strong concrete example of domain-specific skills done properly: the agent learns when to use MCP, when to drop to a Python SDK, and when to call a raw API, while the user stays in natural language. Worth adding because it shows that ""skills"" are not just prompt snippets, but curated execution strategies that hide a multi-tool integration stack behind intent." aiboost_harness,245,Design Primitives > Skills & MCP,You can't whisper at an AI agent,https://stripe.dev/blog/ai-steering-experiments,,,2026,5,text_date,grey,"Stripe's May 2026 study of how agents actually consume SDK and CLI guidance: passive documentation is ignored, while hard steering signals placed in the loaded context — skill files, error messages, CLI prompts — reliably change behavior. The ""if your guidance wasn't in the loaded context, it didn't happen"" principle is a practical design rule for any skill, tool, or SDK that agents are expected t" aiboost_harness,246,Design Primitives > Skills & MCP,Agent Toolkit for AWS,https://github.com/aws/agent-toolkit-for-aws,,,,,,repository,"Official AWS-supported MCP servers, skills, and plugins that let AI agents provision, query, and manage AWS resources through a standardized protocol interface. Worth including as the reference for how a major cloud provider productizes infrastructure access into agent-ready harness primitives rather than leaving teams to hand-roll IAM-scoped tool definitions." aiboost_harness,247,Design Primitives > Skills & MCP,agentic-stack,https://github.com/codejunkie99/agentic-stack,,,,,,repository,"Portable .agent/ folder that externalizes memory, skills, and protocols from any specific coding agent into a cross-tool harness layer. Adapters translate the same configuration into Claude Code's CLAUDE.md, Cursor's rules, OpenCode's AGENTS.md, and more — the first practical answer to harness vendor lock-in at the configuration level." aiboost_harness,248,Design Primitives > Skills & MCP,mcp-agent,https://github.com/lastmile-ai/mcp-agent,,,,,,repository,"Production-grade framework for building agents with MCP: composable workflows, built-in observability, and provider-agnostic model routing. The clearest reference for turning MCP servers from isolated utilities into a coherent agent harness." aiboost_harness,249,Design Primitives > Skills & MCP,vurb.ts,https://github.com/vinkius-labs/vurb.ts,,,,,,repository,"TypeScript framework for building production MCP servers with a ""Presenter"" perception layer that strips undeclared fields, redacts PII, and gates tool visibility by workflow state. Fills a critical gap in the MCP ecosystem: most tooling focuses on consuming servers, while vurb.ts addresses the harness engineering of authoring servers that are safe, governable, and context-aware by default." aiboost_harness,250,Design Primitives > Skills & MCP,SkillOpt,https://github.com/microsoft/SkillOpt,,,,,,repository,"Microsoft's skill optimizer that trains reusable natural-language skills for frozen LLM agents through trajectory-driven edits and validation-gated updates, producing deployable best_skill.md artifacts. The key harness insight is that skills should be treated as optimizable parameters that improve with execution feedback, not static prompt fragments written once and forgotten." aiboost_harness,251,Design Primitives > Skills & MCP,superpowers,https://github.com/obra/superpowers,,,,,,repository,"Agentic skills framework and software-development methodology whose session-start hook loads a meta-skill that makes skill use mandatory and intent-triggered, then chains brainstorming → written plan → subagent-driven TDD execution in isolated git worktrees → review → verification-before-completion. Ships one skill set across 15+ harnesses (Claude Code, Codex, Cursor, Gemini CLI, Copilot CLI, Open" aiboost_harness,252,Design Primitives > Skills & MCP,Antigravity Awesome Skills,https://github.com/sickn33/antigravity-awesome-skills,,,,,,repository,"Installable library of 1,400+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, and more. The largest community-driven skill catalog with an npm installer and role-based bundles — a concrete reference for treating skills as versioned harness artifacts rather than ad-hoc prompts." aiboost_harness,253,Design Primitives > Skills & MCP,agentgateway,https://github.com/agentgateway/agentgateway,,,,,,repository,"Open-source agentic proxy that unifies LLM gateway, MCP gateway, and A2A gateway into a single control plane for managing multi-agent, multi-tool connectivity at scale. Provides drop-in security, observability, and governance for agent-to-LLM, agent-to-tool, and agent-to-agent communication — the missing infrastructure layer between agents and the services they touch." aiboost_harness,254,Design Primitives > Skills & MCP,AIP: A Graph Representation for Learning and Governing Agent Skills,https://arxiv.org/abs/2606.04781,2606.04781,,2026,6,arxiv_id,academic,"June 2026 proposal to replace free-form skill prose with directed execution graphs: discrete steps as nodes backed by deterministic scripts or natural-language descriptions, connected by explicit typed input/output edges and governed by a schema-validated YAML spec. Compiling skills to AIP improved Claude Sonnet's pass rate from 53% to 67% while making skills queryable, auditable, and repairable a" aiboost_harness,255,Design Primitives > Skills & MCP,Ponytail,https://github.com/DietrichGebert/ponytail,,,,,,repository,"Skill that makes coding agents behave like a ""lazy senior dev"": prefer built-in solutions, avoid new dependencies, and write the minimum code that works. Benchmarked on real Claude Code sessions with ~54% fewer lines, ~20% lower cost, and preserved safety guards — a rare harness-level incentive that fights over-engineering rather than just adding capability." aiboost_harness,256,Design Primitives > Skills & MCP,wshobson/agents,https://github.com/wshobson/agents,,,,,,repository,"Cross-harness plugin marketplace that maintains one source-of-truth plugins/ directory and generates harness-native artifacts for Claude Code, Codex CLI, Cursor, OpenCode, Gemini CLI, and GitHub Copilot. It is the clearest practical example of treating reusable agent capabilities as a portable distribution format rather than ad-hoc prompt files." aiboost_harness,257,Design Primitives > Skills & MCP,mgechev/skillgrade,https://github.com/mgechev/skillgrade,,,,,,repository,"CLI that turns agent skill verification into repeatable unit tests: it generates task and grader pairs from a SKILL.md, runs multi-trial evals against Claude, Codex, Gemini, or OpenCode, and reports pass rates with a CI-ready threshold. Fills the gap between shipping a skill and knowing an agent actually discovers and invokes it correctly." aiboost_harness,258,Design Primitives > Skills & MCP,Qwen-MM-Plugins,https://github.com/QwenLM/Qwen-MM-Plugins,,,,,,repository,"Qwen's official multimodal plugin suite packages vision, video, document, 3D, and CAD capabilities as portable skills and MCP servers across Claude Code, Codex, OpenCode, and other harnesses. It shows how to make a text-first coding harness multimodal-native without rebuilding the agent loop." aiboost_harness,259,Design Primitives > Skills & MCP,Agent-Loop-Skills,https://github.com/gaasher/agent-loop-skills,,,,,,repository,"Turns the agent loop itself into portable Agent Skills: 25 verification-gated iteration loops (autoresearch, scientific writing, data analysis, code optimization, red-teaming) that bind to a task at invocation time and iterate propose → run → score → keep/revert until results improve. The clearest example of the ""loop is the skill"" pattern — externalizing iteration control as a cross-harness artif" aiboost_harness,260,Design Primitives > Skills & MCP,Mods overview — Claude Code Docs,https://code.claude.com/docs/en/plugins/mods/overview,,,2026,10,text_date,grey,"Anthropic's October 2026 extension primitive that moves harness customization from outside the loop to inside it: a mod is a TypeScript function that hooks Claude Code events (tool calls, prompts, turns, UI draws) and can observe, rewrite, block, or answer them — something settings hooks, skills, and MCP servers structurally cannot do. Side effects are funneled through a declared $ API so claude p" aiboost_harness,261,Design Primitives > Skills & MCP,Claude Design Skills Tested,https://agentcraft.dev/en/blog/claude-design-skills-tested/,,,,,,grey,"AgentCraft's same-brief test of the 8 most installed Claude design skills plus a no-skill control (Claude Haiku, a landing page and a portfolio, 18 runs, about $0.90, file tools only). Only Anthropic's frontend-design and taste-skill moved the page away from the default on both briefs, while three skills with hundreds of thousands of installs changed little. A worked example of checking a skill ag" aiboost_harness,262,Design Primitives > Skills & MCP,Oh My Android,https://github.com/ateymoori/oh-my-android,,,,,,repository,"Native macOS control panel for the Android Emulator (dark mode, font scale, locale, TalkBack, network throttling, GPS mock, Layout Inspector in dp) with a built-in stdio MCP server (18 tools) that lets Claude Code, Codex, Cursor, and VS Code see and drive the emulator. Gives mobile coding agents a structured way to observe and verify the app they just built instead of guessing from raw adb output." aiboost_harness,263,Design Primitives > Skills & MCP,Firecrawl,https://github.com/firecrawl/firecrawl-mcp-server,,,,,,repository,"MCP server for web search, scraping, and browser interaction that returns pages as Markdown or as JSON matching a schema the agent supplies, instead of raw HTML. Requesting only the fields a task needs through that schema is a practical way to keep web content from flooding the context window." aiboost_harness,264,Design Primitives > Skills & MCP,"MCP servers over stdio: startup time, call latency and failures",https://theagenticweekly.com/research/mcp-servers-latency-and-failures,,,,,,grey,"Agentic Weekly timed 15 MCP server setups over stdio with the official Python SDK (cold start, tools/list, median and p95 call latency) and logged what broke, such as the Fetch server refusing localhost by default and 2.x reference servers needing Node 24. Useful for sizing per-call protocol overhead in a harness." aiboost_harness,268,Design Primitives > Permissions & Authorization,Beyond Permission Prompts,https://www.anthropic.com/engineering/claude-code-sandboxing,,,,,,grey,Structured authorization patterns for agents: how to give agents the right permissions without relying on prompt-level trust. aiboost_harness,269,Design Primitives > Permissions & Authorization,OWASP LLM06:2025 — Excessive Agency,https://genai.owasp.org/llmrisk/llm062025-excessive-agency/,,,,,,grey,"OWASP's authoritative definition of the ""excessive agency"" risk: over-provisioned functions, unnecessary permissions, and missing approval mechanisms. The standard checklist for auditing harness permission scope against principle of least privilege." aiboost_harness,270,Design Primitives > Permissions & Authorization,GitHub Enterprise — Governing Agents,https://wellarchitected.github.com/library/governance/recommendations/governing-agents/,,,2026,4,text_date,repository,"April 2026 GitHub official guide for enterprise agent governance: MCP server registry curation with ruleset-protected configurations, agent environment standardization via copilot-setup-steps.yml, ephemeral runner enforcement, and cloud-agent firewall allowlisting. The most concrete published reference for governing agent fleets at scale without creating bottlenecks." aiboost_harness,271,Design Primitives > Permissions & Authorization,Claude Code Auto Mode: A Safer Way to Skip Permissions,https://www.anthropic.com/engineering/claude-code-auto-mode,,,,,,grey,"Anthropic's engineering post on replacing approval fatigue (users approve 93% of prompts, making approvals meaningless) with a two-stage classifier: fast single-token gate first, chain-of-thought reasoning only on flagged actions. The design decisions — stripping assistant messages to prevent the agent from rationalizing dangerous actions, deny-and-continue recovery instead of halt — are the refer" aiboost_harness,272,Design Primitives > Permissions & Authorization,Claude Agent SDK — Configure Permissions,https://platform.claude.com/docs/en/agent-sdk/permissions,,,,,,grey,"The most concrete reference for harness permission architecture: five-layer evaluation order (hooks → deny rules → permission mode → allow rules → canUseTool), allowedTools/disallowedTools declarative scoping, and four permission modes including dontAsk (deny-by-default for headless agents). The subagent inheritance warning for bypassPermissions alone is worth reading before any multi-agent deploy" aiboost_harness,273,Design Primitives > Permissions & Authorization,Two Different Types of Agent Authorization,https://blog.langchain.com/two-different-types-of-agent-authorization/,,,,,,grey,"Distinguishes on-behalf-of authorization (agent uses end-user credentials, requires cross-channel identity mapping and per-user memory isolation) from fixed-credential authorization (agent owns its own account, requires human-in-the-loop guardrails on high-risk actions). The two models have fundamentally different threat surfaces and determine where authorization enforcement lives in the harness." aiboost_harness,274,Design Primitives > Permissions & Authorization,Authorization and Governance for AI Agents: Runtime Authorization Beyond Identity at Scale,https://techcommunity.microsoft.com/blog/microsoft-security-blog/authorization-and-governance-for-ai-agents-runtime-authorization-beyond-identity/4509161,,,,,,grey,"Microsoft Security's reusable Authorization Fabric combining a Policy Enforcement Point (PEP) and Policy Decision Point (PDP) as a Microsoft Entra-protected endpoint. Every agent calls this fabric before tool execution, receiving a deterministic decision: ALLOW / DENY / REQUIRE_APPROVAL / MASK. Addresses the gap that identity alone (who is this agent?) doesn't answer whether a specific action shou" aiboost_harness,275,Design Primitives > Permissions & Authorization,IETF draft-klrc-aiagent-auth: AI Agent Authentication and Authorization,https://datatracker.ietf.org/doc/draft-klrc-aiagent-auth/,,,2026,3,text_date,grey,"The first IETF standards-track specification for AI agent authentication (March 2026, authors from AWS, OpenAI, Zscaler, Ping Identity, Defakto Security). Builds on WIMSE (Workload Identity in Multi-System Environments) and OAuth 2.0 rather than inventing new protocols — agents get SPIFFE-style identifiers, with delegation via OAuth Token Exchange and DPoP for token binding. Essential reference fo" aiboost_harness,276,Design Primitives > Permissions & Authorization,Nango: Pre-Built Authentication for AI Agents,https://nango.dev,,,,,,grey,"Open-source platform providing pre-built OAuth and API key authentication for 700+ APIs across 30 categories. Automatically refreshes access tokens, provides webhooks when credentials break, and stores tokens securely so agent code never touches secrets. Solves the ""agent needs to call an authenticated API"" problem at scale — the authentication layer that complements Composio's tool wrapping." aiboost_harness,277,Design Primitives > Permissions & Authorization,Agent Vault,https://github.com/Infisical/agent-vault,,,,,,repository,"Infisical's open-source credential broker that sits between AI agents and the APIs they call, injecting real credentials onto outbound requests so agents never possess secrets directly. Eliminates a concrete prompt-injection attack surface — exfiltration of API keys and PATs — by treating credential possession as a harness-layer boundary rather than an agent-side configuration." aiboost_harness,278,Design Primitives > Permissions & Authorization,AgentDoG: A Diagnostic Guardrail Framework for AI Agent Safety and Security,https://arxiv.org/abs/2601.18491,2601.18491,,2026,1,arxiv_id,academic,Three-dimensional risk taxonomy (source/failure-mode/consequence) with fine-grained agentic safety benchmark (ATBench) and diagnostic guardrail models (4B–8B parameters) achieving 91.8% accuracy. Shifts safety monitoring from binary safe/unsafe checks to root-cause diagnosis: why did an action violate constraints? Where did the violation originate? What are the downstream consequences? Essential f aiboost_harness,279,Design Primitives > Permissions & Authorization,Open Agent Passport (OAP): Deterministic Pre-Action Authorization for Autonomous AI Agents,https://arxiv.org/abs/2603.20953,2603.20953,,2026,3,arxiv_id,academic,"Deterministic Pre-Action Authorization for Autonomous AI Agents](https://arxiv.org/abs/2603.20953) — March 2026 open specification and reference implementation that intercepts tool calls synchronously before execution, evaluates them against a declarative policy, and produces a cryptographically signed audit record. Enforces authorization in a median of 53ms; in a live adversarial testbed ($5,000 " aiboost_harness,281,Design Primitives > Permissions & Authorization,Aegis,https://github.com/Justin0504/Aegis,,,,,,repository,"Pre-execution firewall that intercepts, classifies, and blocks agent tool calls before they execute, with a compliance cockpit for real-time monitoring, human-in-the-loop approvals, and a tamper-evident audit trail. The zero-code-change integration makes runtime policy enforcement practical for existing agent deployments." aiboost_harness,282,Design Primitives > Permissions & Authorization,"When ""Do Not"" Is Not Deny: Security Rules in CLAUDE.md vs Built-In Controls",https://arxiv.org/abs/2608.23550,2608.23550,,2026,8,arxiv_id,academic,"Analyzes 481 public CLAUDE.md files and finds only ~4% of natural-language security rules are backed by a matching built-in control, exposing the gap between documented intent and enforced permissions. A concrete reminder that harness instructions are not guardrails unless they map to deterministic enforcement." aiboost_harness,283,Design Primitives > Permissions & Authorization,Do User-Authored Permission Policies Improve Protection Against AI Agent Overreach?,https://arxiv.org/abs/2608.27443,2608.27443,,2026,8,arxiv_id,academic,"August 2026 user study (113 non-technical participants) comparing per-action HITL approval, automated model review, and user-authored allow/ask/never rules. The key harness finding: pre-authored policies blocked ~20 percentage points *less* overreach than per-action approval — users overwhelmingly set ""ask"" as their rule and then approve anyway, so static rule pre-commitment degenerates back into " aiboost_harness,284,Design Primitives > Permissions & Authorization,Connections: Managed Credentials and Per-Caller Identity for Managed Deep Agents,https://www.langchain.com/blog/connections-managed-credentials-and-per-caller-identity-for-managed-deep-agents,,,2026,9,text_date,grey,"LangChain's September 2026 implementation of on-behalf-of authorization as a harness primitive: credentials live in the workspace (never in .env or the build) and are resolved at run time by slug, with user-owned connections resolving per caller so agent actions carry the requesting human's identity rather than a shared service account. The two-axis model (agent-owned vs. user-owned × static secre" aiboost_harness,288,Design Primitives > Memory & State,Building Effective Agents,https://www.anthropic.com/research/building-effective-agents,,,,,,grey,"Covers in-context, external, and procedural memory patterns as harness-level concerns." aiboost_harness,289,Design Primitives > Memory & State,Letta (MemGPT),https://github.com/letta-ai/letta,,,,,,repository,](https://github.com/letta-ai/letta) — The reference architecture for stateful agents: three-tier memory (core / archival / recall) maps directly to harness state management design. Their [agent loop redesign post](https://www.letta.com/blog/letta-v1-agent) is the most thorough public analysis of how memory structure shapes the harness. aiboost_harness,290,Design Primitives > Memory & State,mem0,https://github.com/mem0ai/mem0,,,,,,repository,"Drop-in universal memory layer (YC-backed, AWS Agent SDK's exclusive memory provider) that handles cross-session retention without custom harness-level state management code. Lowest integration cost for production-grade persistent memory." aiboost_harness,291,Design Primitives > Memory & State,Stash,https://github.com/alash3al/stash,,,,,,repository,"Self-hosted persistent memory layer with an 8-stage consolidation pipeline (episodes → facts → relationships → patterns) and built-in MCP server. The critical gap it fills: production-grade cross-session memory without cloud dependencies or complex infrastructure — a single Docker Compose gives you Postgres, pgvector, and background consolidation." aiboost_harness,292,Design Primitives > Memory & State,TencentDB-Agent-Memory,https://github.com/Tencent/TencentDB-Agent-Memory,,,,,,repository,"Tencent's fully local agent memory system with a 4-tier progressive pipeline (Conversation → Atom → Scenario → Persona) and symbolic short-term memory via Mermaid canvases. The benchmark data is striking: 61% token reduction and 51% relative pass-rate improvement on long-horizon tasks, demonstrating that hierarchical memory architecture outperforms flat vector stores for coding agents." aiboost_harness,294,Design Primitives > Memory & State,engram,https://github.com/Gentleman-Programming/engram,,,,,,repository,"Persistent memory for AI coding agents delivered as a single Go binary with SQLite + FTS5 and 18 MCP tools for save, search, session lifecycle, and conflict detection. The agent-agnostic, zero-dependency design makes cross-session memory a local harness primitive rather than a managed cloud service." aiboost_harness,295,Design Primitives > Memory & State,MemPalace,https://github.com/MemPalace/mempalace,,,,,,repository,"Local-first AI memory system that stores conversation history verbatim and retrieves it with semantic search through a structured palace architecture (wings, rooms, drawers). Achieves 96.6% R@5 on LongMemEval with zero LLM calls, making it the best-benchmarked open-source memory layer for agents that need cross-session persistence without cloud dependencies." aiboost_harness,296,Design Primitives > Memory & State,agentmemory,https://github.com/rohitg00/agentmemory,,,,,,repository,"Persistent memory server for coding agents that retrieves only the top-K relevant observations via hybrid BM25 + vector + graph search instead of loading a whole memory file into context (~1,900 vs 22K+ tokens at 240 observations, 92% fewer). Reports 95.2% R@5 on LongMemEval-S with a local embedding model, plus a reproducible eval harness to check that claim. One server serves 20+ agents (Claude C" aiboost_harness,297,Design Primitives > Memory & State,claude-memory-compiler,https://github.com/coleam00/claude-memory-compiler,,,,,,repository,"Turns raw Claude Code sessions into a self-evolving knowledge base: hooks capture every interaction, the Agent SDK extracts decisions and lessons, and an LLM compiler distills them into structured, cross-referenced articles that improve retrieval quality over time. The most concrete open-source implementation of trace-driven memory evolution for coding agents." aiboost_harness,299,Design Primitives > Memory & State,How We Built Agent Builder's Memory System,https://blog.langchain.com/how-we-built-agent-builders-memory-system/,,,,,,grey,"LangChain's engineering account of a COALA-based three-tier memory system (procedural/semantic/episodic) backed by PostgreSQL but exposed to agents as a virtual filesystem. Key harness decisions: human-in-the-loop approval gates every memory write (blocking prompt-injection via malformed writes), validation errors are fed back to the LLM for self-correction, and AGENTS.md serves as the agent's pro" aiboost_harness,300,Design Primitives > Memory & State,When Malicious Instructions Persist: Persistent Memory Poisoning Attack on Harness-Based Agents,https://arxiv.org/abs/2609.13889,2609.13889,,2026,9,arxiv_id,academic,"September 2026 attack paper demonstrating that untrusted content tricked into a coding agent's persistent memory triggers cross-session malicious behavior later: 73.7% injection / 55.5% cross-session attack success on OpenClaw and 66.9% / 81.7% on Claude Code, with benign task performance unchanged so the poisoning stays invisible. The harness takeaway sharpens why the Agent Builder write-up above" aiboost_harness,301,Design Primitives > Memory & State,Building an Agentic Memory System for GitHub Copilot,https://github.blog/ai-and-ml/github-copilot/building-an-agentic-memory-system-for-github-copilot/,,,2026,1,text_date,grey,"GitHub's January 15, 2026 write-up is one of the clearest public discussions of deployed cross-agent memory: repository-scoped memories are shared across coding agent, CLI, and code review, but only after just-in-time verification against the current code state. The core harness lesson is that memory quality is mostly a freshness and invalidation problem — stale, branch-specific memories are often" aiboost_harness,302,Design Primitives > Memory & State,Building Self-Correcting Memory in OpenWiki,https://www.langchain.com/blog/self-correcting-memory-openwiki,,,2026,8,text_date,grey,"LangChain's August 2026 design account of memory that can forget: every claim the agent writes is persisted with versioned code evidence, and when that evidence changes the claim is flagged stale — remaining durably uncertain until re-verified rather than silently trusted. The most concrete published mechanism for memory invalidation: staleness becomes a first-class, inspectable harness state inst" aiboost_harness,303,Design Primitives > Memory & State,MemArchitect: A Policy-Driven Memory Governance Layer,https://arxiv.org/abs/2603.18330,2603.18330,,2026,3,arxiv_id,academic,"Proposes a governance layer that decouples memory lifecycle management (decay, conflict resolution, privacy enforcement) from model weights, directly addressing the ""zombie memory"" problem: outdated facts sitting in the context window that only a harness-level eviction policy — not the model — can remove." aiboost_harness,304,Design Primitives > Memory & State,Codified Context: Infrastructure for AI Agents in a Complex Codebase,https://arxiv.org/abs/2602.20478,2602.20478,,2026,2,arxiv_id,academic,"Production-validated architecture (283 sessions, 108k-line codebase) built on three components: a ""hot-memory constitution"" encoding conventions and multi-agent coordination protocols, 19 domain-specialist agents, and a ""cold-memory knowledge base"" of 34 on-demand specification documents. The empirical data distinguishes what must live in always-on context from what should be retrieved on demand —" aiboost_harness,305,Design Primitives > Memory & State,Facts as First Class Objects: Knowledge Objects for Persistent LLM Memory,https://arxiv.org/abs/2603.17781,2603.17781,,2026,3,arxiv_id,academic,"Identifies three production failure modes of in-context memory at scale: capacity overflow at ~8,000 facts, 60% fact destruction during compaction, and 54% behavioral drift from constraint erosion across cascaded summarizations. Proposes Knowledge Objects (hash-addressed discrete fact tuples) achieving 100% accuracy at 252× lower cost than in-context storage — the quantitative case for moving pers" aiboost_harness,306,Design Primitives > Memory & State,Governance Decay: How Context Compaction Silently Erases Safety Constraints in Long-Horizon LLM Agents,https://arxiv.org/abs/2606.22528,2606.22528,,2026,6,arxiv_id,academic,"June 2026 paper establishing context compaction as a first-class governance surface: the ConstraintRot benchmark (1,323 episodes, seven model families) shows constraint-violation rates rising from 0% with the policy in full context to 30% after compaction (59% for some models), and a Compaction-Eviction Attack that biases the summarizer to drop legitimate policies defeats every evaluated model. Th" aiboost_harness,307,Design Primitives > Memory & State,Recoverability Has a Law: The ERR Measure for Tool-Augmented Agents,https://arxiv.org/abs/2601.22352,2601.22352,,2026,1,arxiv_id,academic,Formal framework for measuring how well agents recover from tool failures. Defines Expected Recovery Regret (ERR) as a metric for harness design: the cost of recovering from stochastic failures in downstream tasks. Critical for assessing reliability of production harnesses where tool calls occasionally fail but agents must continue functioning. aiboost_harness,308,Design Primitives > Memory & State,MAGMA: Multi-Graph Agentic Memory Architecture,https://arxiv.org/abs/2601.03236,2601.03236,,2026,1,arxiv_id,academic,"Represents agent memory across four orthogonal semantic, temporal, causal, and entity graphs, enabling policy-guided retrieval over relational views. Outperforms MemGPT on long-horizon reasoning benchmarks by 18.5% accuracy improvement. The multi-graph abstraction lets harness engineers compose different retrieval strategies for different task phases — a concrete architecture for memory that scale" aiboost_harness,309,Design Primitives > Memory & State,GAAMA: Graph Augmented Associative Memory for Agents,https://arxiv.org/abs/2603.27910,2603.27910,,2026,3,arxiv_id,academic,Hybrid memory system blending graph traversal with semantic similarity through additive scoring; graph augmentation improves retrieval over embedding-only approaches for long-horizon reasoning. Practical alternative to full multi-graph systems when adding structure to existing vector-based memory is sufficient. aiboost_harness,310,Design Primitives > Memory & State,Graph-Native Cognitive Memory for AI Agents: Formal Belief Revision Semantics for Versioned Memory Architectures,https://arxiv.org/abs/2603.17244,2603.17244,,2026,3,arxiv_id,academic,"Formal semantics for versioned memory graphs with belief revision operations, enabling agents to maintain coherent evolving world models through multi-turn reasoning. Addresses the hard problem of inconsistency resolution in long-lived agent memory: when new information contradicts prior beliefs, how should the agent update its knowledge base?" aiboost_harness,311,Design Primitives > Memory & State,Continual learning for AI agents,https://blog.langchain.com/continual-learning-for-ai-agents/,,,2026,4,text_date,grey,"LangChain's April 2026 framing of agent learning as three distinct layers: model weights, harness behavior, and contextual memory. Essential for designing memory systems that don't just store facts but actually improve agent performance over time through trace-driven harness and context updates." aiboost_harness,312,Design Primitives > Memory & State,cognee,https://github.com/topoteretes/cognee,,,,,,repository,"Open-source memory platform with a hybrid graph-vector-relational poly-store that lets agents recall facts through both semantic similarity and structured graph traversal — the practical middle ground between flat vector stores and full multi-graph research systems. The self-improving pipeline reweights edges from agent feedback, and the 14-tool MCP server makes it a drop-in memory primitive for p" aiboost_harness,313,Design Primitives > Memory & State,Hindsight,https://github.com/vectorize-io/hindsight,,,2026,6,text_date,repository,"Agent memory system organized around three explicit operations—retain, recall, and reflect—with semantic, keyword, graph, and temporal retrieval plus an MCP server. The June 2026 release and production usage make it a concrete reference for turning cross-session persistence from passive storage into an active learning layer inside the harness." aiboost_harness,314,Design Primitives > Memory & State,ClawVM: Harness-Managed Virtual Memory for Stateful Tool-Using LLM Agents,https://arxiv.org/abs/2604.10352,2604.10352,,2026,4,arxiv_id,academic,"Applies virtual-memory semantics to agent context management, treating the context window as working memory with typed pages, minimum-fidelity invariants, and validated writeback at lifecycle boundaries. A concrete reference for making residency and durability auditable harness-level concerns rather than best-effort side effects." aiboost_harness,315,Design Primitives > Memory & State,MAGE: Memory as Agent-Guided Exploration,https://arxiv.org/abs/2606.06090,2606.06090,,2026,6,arxiv_id,academic,"June 2026 proposal to treat long-horizon memory as execution-state management rather than semantic retrieval: a hierarchical state tree preserves trajectories, enables rollback, and constructs working state from the active root-to-leaf path. Improves task success by 7.8–20.4 percentage points while cutting token consumption 55.1%, making it a concrete reference for memory architectures where execu" aiboost_harness,316,Design Primitives > Memory & State,deja-vu,https://github.com/vshulcz/deja-vu,,,,,,repository,"Indexes coding-agent sessions already written to disk and serves them back over MCP with no LLM calls, embeddings, or API keys. The zero-dependency binary solves the memory cold-start problem for cross-session persistence and demonstrates that agent memory can be built on existing filesystem traces rather than a separate learning pipeline." aiboost_harness,317,Design Primitives > Memory & State,trajectory,https://github.com/letta-ai/trajectory,,,2026,7,text_date,repository,"Letta's July 2026 library that normalizes native session transcripts from 15+ harnesses (Claude Code, Codex, Cursor, OpenCode, Pi, Gemini CLI, OpenHands, and more) into one validated, model-ready record format. Fills a real infrastructure gap — every runtime logs the same concepts (messages, reasoning, tool calls, results) in incompatible formats — and is explicitly designed for *agent* consumptio" aiboost_harness,318,Design Primitives > Memory & State,Harness the Memory: A Holistic Evaluation of Memory Substrates in Memory Agents,https://arxiv.org/abs/2608.15008,2608.15008,,2026,8,arxiv_id,academic,"August 2026 controlled study running every common memory substrate (dense/sparse indices, text records, structural and hierarchical stores, refinement-based memory, parametric updates) through one unified harness: 26 metrics, 3 backbone models, 4 benchmark suites. The headline finding — no substrate consistently dominates, and broad retrieval helps long-context factual QA while actively harming se" aiboost_harness,319,Design Primitives > Memory & State,"Who Remembers What? Multi-Granularity Memory Extraction for Multi-User, Multi-Agent Interaction",https://arxiv.org/abs/2611.10439,2611.10439,,2026,11,arxiv_id,academic,"November 2026 paper (MixUA) on memory writes in harnesses shared by many users and heterogeneous agents: each lesson distilled from trajectories is assigned a scope (user-agent pair, user, agent, or global) based on who it helps and what it could disclose. A concrete reference for treating memory scoping and access control as a harness decision rather than a single flat shared store." aiboost_harness,323,Design Primitives > Task Runners & Orchestration,LightDAG (LightAgent),https://github.com/wanxingai/LightAgent/blob/v0.11.0/docs/lightdag.md,,,,,,repository,"](https://github.com/wanxingai/LightAgent/blob/v0.11.0/docs/lightdag.md) — Single-host dynamic task graphs with SQLite persistence, bounded concurrent workers, and restart recovery through leases and fencing. Its application-owned verifiers gate task completion and immutable artifact publication, providing a concrete pattern for combining recoverable scheduling with explicit acceptance checks." aiboost_harness,324,Design Primitives > Task Runners & Orchestration,Harness Engineering,https://openai.com/index/harness-engineering/,,,,,,grey,"How task runners fit into the harness: queueing, parallelism, and progress reporting." aiboost_harness,325,Design Primitives > Task Runners & Orchestration,Building a C Compiler with a Team of Parallel Claudes,https://www.anthropic.com/engineering/building-c-compiler,,,,,,grey,"Anthropic's account of coordinating 16 Claude instances in parallel on a shared git repo without a central orchestrator: agents claim tasks via files in current_tasks/, git forces collision resolution naturally, and a continuous restart loop spawns fresh sessions that resume where predecessors left off. Key harness lesson: verbose test output pollutes agent context — the feedback loop must emit on" aiboost_harness,326,Design Primitives > Task Runners & Orchestration,LiteLLM,https://github.com/BerriAI/litellm,,,,,,repository,"Unified proxy and SDK that routes to 100+ LLM providers behind a single OpenAI-compatible interface, with a Router handling retry/fallback across deployments, per-project cost and rate-limit tracking, and OTEL callback integrations. The right infrastructure layer when your harness needs provider resilience (automatic failover on 429/500 errors), budget guardrails, or the ability to swap models wit" aiboost_harness,327,Design Primitives > Task Runners & Orchestration,LangGraph,https://github.com/langchain-ai/langgraph,,,,,,repository,"Graph-based state machine framework for multi-agent harnesses: models supervisor/subagent topologies, error-recovery branches, and checkpoint persistence as first-class primitives. The most widely adopted harness orchestration layer in production." aiboost_harness,328,Design Primitives > Task Runners & Orchestration,OpenAI Agents SDK,https://github.com/openai/openai-agents-python,,,,,,repository,Lightweight multi-agent framework built around handoffs and guardrails; the production successor to Swarm. Complements LangGraph for harnesses where delegation patterns are simpler than full graph orchestration. aiboost_harness,329,Design Primitives > Task Runners & Orchestration,Codex SDK,https://developers.openai.com/codex/sdk,,,,,,grey,"OpenAI's official SDK for programmatically controlling local Codex agents from TypeScript or Python: start threads, run prompts, resume sessions, and choose sandbox presets (read-only, workspace-write, full-access) so Codex can be wired into CI/CD or custom agent surfaces instead of remaining locked inside the CLI." aiboost_harness,330,Design Primitives > Task Runners & Orchestration,Codex as a platform: build on the open agent harness,https://developers.openai.com/blog/codex-as-a-platform,,,2026,9,text_date,grey,"OpenAI's September 2026 guide to treating the open-source Codex harness as reusable infrastructure rather than a product to adopt wholesale: your application owns the interface, business context, MCP tools, and approval gates, while the app-server owns the agent loop and sandboxed execution. The clearest first-party articulation of the division of labor every embedded-agent harness must decide — a" aiboost_harness,331,Design Primitives > Task Runners & Orchestration,Google ADK,https://github.com/google/adk-python,,,,,,repository,"Google's code-first agent framework with built-in multi-agent orchestration, tool registration, session state, and eval pipeline. Its Runner and AgentTool patterns are the reference implementation for wrapping sub-agents as tools in a larger harness." aiboost_harness,332,Design Primitives > Task Runners & Orchestration,strands-agents/harness-sdk,https://github.com/strands-agents/harness-sdk,,,,,,repository,"AWS's open-source, model-driven agent SDK that treats the loop, tool binding, and guardrails as first-class primitives. Supports Bedrock, Anthropic, OpenAI, Gemini, and Ollama with native MCP, built-in observability, and multi-agent patterns — the missing AWS open-source harness framework alongside Bedrock AgentCore." aiboost_harness,333,Design Primitives > Task Runners & Orchestration,"Build Long-running AI agents that pause, resume, and never lose context with ADK",https://developers.googleblog.com/build-long-running-ai-agents-that-pause-resume-and-never-lose-context-with-adk/,,,2026,5,text_date,grey,"Google's May 2026 guide to production agents that survive days-long idle periods: DatabaseSessionService for persistent sessions, webhook-triggered state_delta resumption that lets containers scale to zero, and explicit state machines instead of dumping raw JSON into vector databases. Complements Anthropic's managed-agents architecture with a cloud-native deployment perspective." aiboost_harness,334,Design Primitives > Task Runners & Orchestration,Build Cross-Language Multi-Agent Team with Google's Agent Development Kit and A2A,https://developers.googleblog.com/en/build-cross-language-multi-agent-team-with-google-agent-development-kit-and-a2a/,,,2026,6,text_date,grey,"Google's June 2026 production pattern decomposing monolithic agents into cross-language specialists: a Python LLM extraction agent delegates to a Go deterministic validator via A2A, with shared session-state checkpoints and fail-safe routing to manual review. A concrete example of using harness-level orchestration—rather than a bigger model—to solve context degradation and blast radius in multi-ag" aiboost_harness,335,Design Primitives > Task Runners & Orchestration,AutoGen,https://github.com/microsoft/autogen,,,,,,repository,"Microsoft's multi-agent conversation framework with a complete AgentChat layer covering agent loop, tool integration, termination conditions, and human-in-the-loop. The most comprehensive open-source reference for large-scale multi-agent harness design." aiboost_harness,336,Design Primitives > Task Runners & Orchestration,CrewAI,https://github.com/crewAIInc/crewAI,,,,,,repository,"Dual-layer harness orchestration: Crew handles autonomous agent delegation, Flow provides event-driven deterministic control (branching + shared Pydantic state). The clearest open-source example of mixing autonomous and scripted execution in the same harness." aiboost_harness,337,Design Primitives > Task Runners & Orchestration,Pydantic AI v2,https://github.com/pydantic/pydantic-ai,,,2026,6,text_date,repository,"June 2026 harness-first redesign built around the Capability primitive: a single composable unit bundling instructions, tools, lifecycle hooks, and model settings. The split between a small stable core and a fast-moving pydantic-ai-harness lets capabilities graduate as they prove essential, while deferred loading keeps unused tools out of the context window." aiboost_harness,338,Design Primitives > Task Runners & Orchestration,LangGraph 2.0 Release,https://github.com/langchain-ai/langgraph,,,,,,repository,"Major 2026 release codifying three years of production orchestration patterns: type-safe streaming, Deploy CLI for managed hosting, and unified agent primitives (Router, Supervisor, Subagent) eliminating the need to hand-roll coordination logic. The persistence layer support for checkpoint-resume recovery is the reference design for resumable multi-day agent tasks." aiboost_harness,339,Design Primitives > Task Runners & Orchestration,OmniRoute: Multi-Provider LLM Gateway,https://github.com/diegosouzapw/OmniRoute,,,,,,repository,"Intelligent routing across multiple LLM providers with load balancing, intelligent fallbacks, rate limiting, and response caching. Achieves 40–60% token cost reduction through smart model routing (cheap models for simple tasks, capable models for complex reasoning). Essential infrastructure for harnesses operating under strict cost budgets where model selection is a per-turn decision." aiboost_harness,340,Design Primitives > Task Runners & Orchestration,OpenSquilla,https://github.com/opensquilla/opensquilla,,,,,,repository,"Token-efficient microkernel agent with an on-device SquillaRouter that dispatches each turn to the cheapest capable model, a unified turn loop across CLI/Web/chat, and persistent memory with on-device embeddings. Demonstrates that harness-level routing and loop optimization — not just model scale — are the binding levers for intelligence density." aiboost_harness,341,Design Primitives > Task Runners & Orchestration,Scaling Managed Agents: Decoupling the Brain from the Hands,https://www.anthropic.com/engineering/managed-agents,,,,,,grey,"Anthropic's production architecture for separating three stateless components — the ""brain"" (Claude + harness), ""hands"" (sandboxes/tools), and ""session"" (append-only event log) — enabling independent failure and replacement of each. Crash recovery via session replay (wake(sessionId) + getEvents()) and on-demand container provisioning cut p50 TTFT by ~60% and p95 by over 90%. The reference design f" aiboost_harness,342,Design Primitives > Task Runners & Orchestration,"Microsoft Agent Framework at BUILD 2026: Agent Harness, Hosted Agents, CodeAct, and more",https://devblogs.microsoft.com/agent-framework/microsoft-agent-framework-at-build-2026-announce/,,,2026,6,text_date,grey,"Microsoft's June 2026 expansion folds previously scattered harness primitives into a single runtime layer: automatic context compaction, instruction merging, todo tracking, background agents, and a ToolApprovalAgent. The CodeAct execution mode is particularly notable—agents emit a short Python program that calls tools in one sandbox run, cutting latency by 52% and tokens by 64%." aiboost_harness,343,Design Primitives > Task Runners & Orchestration,Microsoft Agent Framework 1.0,https://devblogs.microsoft.com/agent-framework/microsoft-agent-framework-version-1-0/,,,2026,4,text_date,grey,"Production-ready 1.0 release (April 2026) unifying Semantic Kernel and AutoGen into a single framework with graph-based orchestration, middleware pipeline for intercepting every execution stage, and declarative YAML agent definitions. DevUI provides a browser-based debugger for visualizing agent execution, message flows, and tool calls in real time. Multi-provider support (Azure OpenAI, Anthropic," aiboost_harness,344,Design Primitives > Task Runners & Orchestration,Conductor,https://github.com/microsoft/conductor,,,,,,repository,"Microsoft’s open-source YAML-first CLI for deterministic multi-agent orchestration: Jinja2 routing, static parallel groups, and per-agent model overrides across Claude and Copilot with zero token overhead on the orchestration layer itself. Treats workflows as version-controlled, diffable infrastructure rather than runtime-discovered topology." aiboost_harness,345,Design Primitives > Task Runners & Orchestration,Shopify Roast,https://github.com/Shopify/roast,,,,,,repository,"Shopify's open-source Ruby DSL for structured AI workflows that interleaves deterministic steps (shell commands, Ruby code) with agentic steps via Claude Code or Pi. Its ""non-determinism is the enemy of reliability"" philosophy makes it a concrete reference for building reproducible, version-controlled harnesses where guardrails are structural rather than prompt-level." aiboost_harness,346,Design Primitives > Task Runners & Orchestration,AgentScope Runtime,https://github.com/agentscope-ai/agentscope-runtime,,,,,,repository,"Production-ready open-source runtime focused on two pieces many agent frameworks leave underspecified: secure sandbox execution and durable agent serving. The ""Agent as API"" model, async sandbox types, and built-in state/sandbox lifecycle management make it one of the few 2026 projects tackling runtime concerns directly instead of stopping at orchestration abstractions." aiboost_harness,347,Design Primitives > Task Runners & Orchestration,AgentScope Java,https://github.com/agentscope-ai/agentscope-java,,,,,,repository,"Production-ready Java framework for distributed, enterprise-grade agents with workspace sandboxing, permission-gated tool calls, AOP-style middleware, and cross-replica session recovery. The Java complement to AgentScope Runtime for teams that need a typed, JVM-native harness stack." aiboost_harness,348,Design Primitives > Task Runners & Orchestration,Orchestrating Ambient Agents with Temporal,https://temporal.io/blog/orchestrating-ambient-agents-with-temporal,,,,,,grey,"Temporal.io's harness infrastructure for persistent agent workflows with native agentic handshake protocol for secure deadline-aware calendar negotiation between autonomous agents. Brings distributed systems best practices (durability, retry semantics, activity monitoring) to agent orchestration, enabling agents to handle long-running tasks that outlast any single HTTP request or session." aiboost_harness,349,Design Primitives > Task Runners & Orchestration,Vercel AI SDK,https://github.com/vercel/ai,,,,,,repository,"The leading TypeScript toolkit for building AI agents (20M+ monthly downloads, 25+ provider integrations). AI SDK 6 introduced a first-class Agent abstraction with ToolLoopAgent for production-ready tool execution loops, DevTools for local debugging, full MCP support, and type-safe UI streaming. The unified API across OpenAI, Anthropic, Google, and AWS Bedrock makes it the default choice for TypeS" aiboost_harness,351,Design Primitives > Task Runners & Orchestration,Mastra,https://github.com/mastra-ai/mastra,,,,,,repository,"TypeScript-native agent framework (from the Gatsby team) with 22K+ stars and 300K+ weekly npm downloads. Connects to 40+ providers through one standard interface, with built-in workflows, RAG pipelines, and agent orchestration. The @mastra/deployer handles serverless deployment, and the eval system supports LLM-as-judge out of the box. The strongest alternative to Vercel AI SDK for teams that need" aiboost_harness,352,Design Primitives > Task Runners & Orchestration,Flue,https://github.com/withastro/flue,,,,,,repository,"Astro's TypeScript-native agent harness that treats an agent as a function composed of model, sandbox, skills, tools, and MCP servers. The deploy-anywhere runtime (Node, Cloudflare Workers, GitHub Actions, Daytona) and first-class durability make it a practical reference for building autonomous agents rather than chatbots." aiboost_harness,353,Design Primitives > Task Runners & Orchestration,open-multi-agent,https://github.com/JackChen-me/open-multi-agent,,,,,,repository,"TypeScript-native multi-agent orchestration that automatically decomposes a natural-language goal into a task DAG, parallelizes independent nodes, and synthesizes results. With only three runtime dependencies, built-in MCP support, token budgets, retries, and context compaction, it is the lightest production-grade harness layer for Node.js backends." aiboost_harness,354,Design Primitives > Task Runners & Orchestration,The next evolution of the Agents SDK,https://openai.com/index/the-next-evolution-of-the-agents-sdk/,,,2026,4,text_date,grey,"OpenAI's April 2026 update adding native sandbox execution, configurable memory, and sandbox-aware orchestration to the Agents SDK. The shift toward a ""model-native harness"" that aligns execution patterns with how frontier models actually perform best is a reference design for SDK-level harness evolution." aiboost_harness,355,Design Primitives > Task Runners & Orchestration,Symphony,https://github.com/openai/symphony,,,,,,repository,"OpenAI's orchestration layer for moving from supervising coding agents to managing work: it monitors an issue tracker, creates isolated per-issue workspaces, and surfaces proof-of-work artifacts (CI status, review feedback, walkthrough videos) as the handoff signal. The spec deliberately leaves sandbox and approval policies implementation-defined so teams can match trust posture to environment, an" aiboost_harness,356,Design Primitives > Task Runners & Orchestration,Harmonist,https://github.com/GammaLabTechnologies/harmonist,,,,,,repository,"Drop-in multi-agent framework where protocol enforcement is a mechanical gate, not a prompt request: IDE-level hooks check every code-changing turn for required reviewers, memory updates, and supply-chain integrity before the turn can complete. Built on stdlib Python with zero dependencies, it demonstrates how to build deterministic harness constraints that even frontier models cannot override." aiboost_harness,357,Design Primitives > Task Runners & Orchestration,Hive,https://github.com/aden-hive/hive,,,,,,repository,"YC-backed production harness that compiles multi-agent objectives into deterministic execution DAGs with state persistence, crash recovery, cost enforcement, and human-in-the-loop oversight. The most complete open-source reference for running agent workloads in production rather than demos." aiboost_harness,358,Design Primitives > Task Runners & Orchestration,thClaws,https://github.com/thClaws/thClaws,,,,,,repository,"Native-Rust agent harness platform with four surfaces (GUI, CLI, web, one-shot), multi-provider routing, and sovereign-by-design architecture. The most complete open-source reference for a locally-run, offline-capable harness that unifies MCP, skills, AGENTS.md, hooks, and session resumption in a single binary." aiboost_harness,359,Design Primitives > Task Runners & Orchestration,sandcastle,https://github.com/mattpocock/sandcastle,,,,,,repository,"TypeScript-native orchestration for sandboxed coding agents that treats provider-agnostic isolation (Docker, Podman, or Vercel Firecracker microVMs) as a primitive, not a framework. The built-in review-pipeline and parallel-AFK-agent patterns demonstrate how lightweight harness layers can enforce safety without the complexity of full graph orchestration." aiboost_harness,360,Design Primitives > Task Runners & Orchestration,bernstein,https://github.com/sipyourdrink-ltd/bernstein,,,,,,repository,"Deterministic scheduler for 40+ CLI coding agents running in parallel git worktrees with an HMAC-signed audit chain, signed agent cards, and per-artefact lineage. The zero-LLM coordination loop and tamper-evident audit trail make it the only open-source orchestrator designed for compliance-sensitive agent fleets." aiboost_harness,361,Design Primitives > Task Runners & Orchestration,waggle,https://github.com/modiqo/waggle,,,,,,repository,"Attributed, resolvable artifact references for agent handoffs: a ~30-byte token replaces pasted context, letting each consumer resolve only the projection or slice it needs under byte budgets while propagating corrections to every holder. The first concrete primitive that turns multi-agent handoffs from a context-copy problem into a reference-and-projection problem." aiboost_harness,362,Design Primitives > Task Runners & Orchestration,Orca,https://github.com/stablyai/orca,,,,,,repository,"The AI orchestrator for running Codex, Claude Code, OpenCode, and Pi side-by-side in isolated git worktrees. It turns fleet-of-agents execution into a polished desktop IDE with a mobile companion, making parallel agent harness design accessible beyond shell-scripting teams." aiboost_harness,363,Design Primitives > Task Runners & Orchestration,LoopX,https://github.com/huangruiteng/loopx,,,,,,repository,"Provider-neutral control plane that sits above existing harnesses (Codex, Claude Code, Cursor) and keeps long-horizon objectives, gates, todos, evidence, and quota stable across bounded agent turns. The local-first state kernel makes multi-day work reviewable, restartable, and handoff-safe without replacing the runtime underneath." aiboost_harness,364,Design Primitives > Task Runners & Orchestration,HarnessRouter,https://github.com/HarnessRouter/harnessrouter,,,,,,repository,"Self-hosted unified API for running Codex, Claude Code, DeepSeek Harness, Pi, and other terminal agents through the open Unified Harness Protocol (UHP), with sessions, streaming, file access, cancellation, and failure handling. Worth including because harness fragmentation is becoming a real operations problem — this is the clearest open-source attempt to make agent runtimes interchangeable behind" aiboost_harness,365,Design Primitives > Task Runners & Orchestration,Scion,https://github.com/GoogleCloudPlatform/scion,,,2026,4,text_date,repository,"Google's experimental ""hypervisor for agents"" (April 2026, Apache-2.0): orchestrates Claude Code, Gemini CLI, Codex, and OpenCode as isolated concurrent processes, each with its own container, git worktree, credentials, and network policy. The core harness insight is *isolation over coordination* — let agents run fully autonomous inside their containers and enforce guardrails at the infrastructure" aiboost_harness,366,Design Primitives > Task Runners & Orchestration,Sno Station,https://github.com/sno-ai/sno-station,,,,,,repository,"Local runtime with no daemon that makes Codex, Claude Code, Cursor, Hermes Agent and OpenClaw work as one team: sno reach pair puts a builder and a challenger on a task, using two vendors when you are signed in to both, and work moves to another agent at a usage limit. Worth including as an open-source reference for cross-vendor review and handoff between harnesses you already run, rather than a n" aiboost_harness,367,Design Primitives > Task Runners & Orchestration,Orchestrator,https://github.com/OrchestratorInc/agent-orchestrator,,,,,,repository,"Open-source desktop app (macOS, Windows, Linux, with iPhone and Android companions) where an orchestrator agent splits work across workers, each in its own git worktree, running Claude Code, Codex and 35+ other coding agents. Worth including for how it closes the verification loop: failed CI checks, review comments and merge conflicts are routed back to the worker that owns the PR, so fixes do not" aiboost_harness,368,Design Primitives > Task Runners & Orchestration,Kraft,https://github.com/itsOmidKarami/kraft,,,,,,repository,"Local orchestrator that keeps the process outside the agent: a YAML chain file sets the order of agent, command and gate nodes, tests and CI run in nodes of their own, a separate reviewer decides whether a fix is done, and attempts, time and spend are capped. Worth including as a worked example of human gates in a coding-agent harness: each work item runs in its own git worktree and stops for a pe" aiboost_harness,374,Design Primitives > Verification & CI Integration,Demystifying Evals for AI Agents,https://www.anthropic.com/engineering/demystifying-evals-for-ai-agents,,,,,,grey,"How to build verification into the harness loop, not just as a post-hoc eval." aiboost_harness,375,Design Primitives > Verification & CI Integration,promptfoo,https://github.com/promptfoo/promptfoo,,,,,,repository,"YAML-driven LLM testing framework with LLM-as-judge, assertion DSL, and native CI integration. The most practical tool for adding agent output regression tests to a PR pipeline without writing a test harness from scratch." aiboost_harness,376,Design Primitives > Verification & CI Integration,AgentBench,https://github.com/THUDM/AgentBench,,,,,,repository,"Multi-environment agent benchmark (OS, DB, web, code) with a structured eval pipeline. Worth studying for its environment isolation design and task definition format when building custom eval environments for your harness." aiboost_harness,377,Design Primitives > Verification & CI Integration,Testing Agent Skills Systematically with Evals,https://developers.openai.com/blog/eval-skills,,,,,,grey,"OpenAI's framework for skill regression testing: four eval dimensions (outcome, process, style, efficiency goals), JSONL trace capture for deterministic checks (command sequences, token budgets, repo cleanliness), then rubric-based grading only where deterministic checks don't suffice. The layering principle — add expensive LLM-as-judge checks only where they reduce meaningful risk — is the most a" aiboost_harness,378,Design Primitives > Verification & CI Integration,Agent Evaluation Readiness Checklist,https://blog.langchain.com/agent-evaluation-readiness-checklist/,,,,,,grey,"A 33-item checklist covering the full evaluation lifecycle: error taxonomy, three-level granularity (single-step → trace → multi-turn thread), grader specialization, and CI integration. Key insight: capability evals (low pass rate, improvement target) and regression evals (near-100%, protection target) must be separated — mixing them produces wrong prioritization decisions." aiboost_harness,379,Design Primitives > Verification & CI Integration,Evaluating Skills,https://blog.langchain.com/evaluating-skills/,,,,,,grey,"LangChain's methodology for benchmarking agent skills in Docker-sandboxed environments. Key empirical findings: Claude Code achieved 82% task completion with curated skills vs. 9% without, and consolidating to ≤12 skills improved accuracy over sprawling skill sets. The baseline-vs-skills comparison design with bugfix tasks and clear outcome metrics is the template for systematic skill coverage tes" aiboost_harness,380,Design Primitives > Verification & CI Integration,AgentAssay: Token-Efficient Regression Testing for Non-Deterministic Agent Workflows,https://arxiv.org/abs/2603.02601,2603.02601,,2026,3,arxiv_id,academic,Addresses agent CI's core problem: binary pass/fail is useless for non-deterministic workflows. Behavioral fingerprinting detects 86% of regressions vs. 0% with binary testing; stochastic PASS/FAIL/INCONCLUSIVE verdicts grounded in hypothesis testing cut token costs 78%. Trace-first offline mode runs regression checks against production traces at zero additional inference cost. aiboost_harness,381,Design Primitives > Verification & CI Integration,Agentic Harness for Real-World Compilers: A Case Study in Specialized Tool Design,https://arxiv.org/abs/2603.20075,2603.20075,,2026,3,arxiv_id,academic,"Demonstrates how harness specialization (llvm-autofix) for a narrow domain (compiler bug fixes) achieves better results than general-purpose coding agents. The tool design patterns — exposing compiler error messages directly, bounding search depth by compilation cost — are transferable to any domain where cost-based pruning and specialized feedback loops make the difference between failure and suc" aiboost_harness,382,Design Primitives > Verification & CI Integration,Eval-Driven Development: Build and Evaluate Reliable AI Agents,https://developers.redhat.com/articles/2026/03/23/eval-driven-development-build-evaluate-ai-agents,,,2026,3,text_date,grey,Red Hat's eight-stage evaluation maturity progression from manual CLI testing to cost-aware continuous monitoring (March 2026). Uses DeepEval with 15 custom ConversationalGEval metrics and LLM-as-judge; key finding: evaluator model capability matters significantly — llama-3-3-70b caught all known failures while smaller models missed 4–5 cases. The $0.64/run cost estimate and self-hosted evaluator aiboost_harness,383,Design Primitives > Verification & CI Integration,"Agent Evaluation Framework 2026: Metrics, Rubrics & Benchmarks",https://galileo.ai/blog/agent-evaluation-framework-metrics-rubrics-benchmarks,,,2026,2,text_date,grey,"Comprehensive framework combining multi-environment baselines (AgentBench), domain-specific benchmarks (Terminal Bench 2.0, WebArena, SWE-bench Verified), and industry standards (NIST AI Agent Standards Initiative, February 2026). Provides reference metrics and rubrics for evaluating coding agents, chatbots, and specialized agents across dimensions (correctness, efficiency, safety). Essential for " aiboost_harness,384,Design Primitives > Verification & CI Integration,The 2025 AI Agent Index: Documenting Technical and Safety Features of Deployed Agentic AI Systems,https://arxiv.org/abs/2602.17753,2602.17753,,2026,2,arxiv_id,academic,"Systematic index analyzing deployed agent safety documentation, guardrails, and third-party testing across 30 production systems; identifies critical gaps in agentic safety disclosure and documentation. Useful as a checklist for production harness safeguards before deployment." aiboost_harness,385,Design Primitives > Verification & CI Integration,sentrux,https://github.com/sentrux/sentrux,,,,,,repository,"Real-time architectural sensor that closes the feedback loop for coding agents: scans codebases, scores structural health, and surfaces degradation via MCP so agents can self-correct before entropy compounds. The gate --save / gate pair makes architectural regression detection CI-friendly — filling the verification gap between linters (style) and tests (behavior) for code produced by autonomous ha" aiboost_harness,386,Design Primitives > Verification & CI Integration,Driving the Agent Quality Flywheel from Your Coding Agent,https://developers.googleblog.com/en/driving-the-agent-quality-flywheel-from-your-coding-agent/,,,2026,6,text_date,grey,"Google's June 2026 account of automating the eval-optimize loop for coding agents: independent AutoRaters grade agent outputs so the optimizer cannot game its own metrics, and custom rubrics isolate specific behaviors like stale-message echoing that blended scores miss. Shows how to close the verification loop continuously against both synthetic dev scenarios and production traces." aiboost_harness,387,Design Primitives > Verification & CI Integration,mcp-test-harness,https://github.com/vaquarkhan/mcp-test-harness,,,,,,repository,"Pytest-style testing framework for MCP servers with protocol-aware assertions, snapshot tests, and CI-ready reports across stdio/SSE/HTTP transports. Fills the gap between shipping an MCP server and trusting it in production by turning tool/schema regressions into ordinary test failures." aiboost_harness,388,Design Primitives > Verification & CI Integration,shadcn-ui/lint,https://github.com/shadcn-ui/lint,,,2026,9,text_date,repository,"September 2026 agent-first linter for Tailwind design systems: you encode design-system contracts as programmable lint rules, and the errors carry your components' sizes, variants, and theme file paths so agents self-correct in one round instead of guessing. Validated across 150+ task runs (published evals: nearly all tasks reach zero violations in a single correction round, and lint feedback cost" aiboost_harness,389,Design Primitives > Verification & CI Integration,"SLCR (Sliceable, Linked, Checkable Requirements)",https://github.com/wack/slcr,,,,,,repository,"](https://github.com/wack/slcr) — An open requirements format and CLI that maps each requirement’s acceptance criteria to validators (tests, linters, ast-grep rules), so slcr check can report in CI which requirements are satisfied rather than relying on the agent’s own claim of done. Agents get per-requirement slices containing only the dependencies and glossary terms a task needs, instead of the " aiboost_harness,393,Design Primitives > Observability & Tracing,AEGIS,https://github.com/antropos17/Aegis,,,,,,repository,"Windows-first desktop monitor for local agent processes, configured file activity, TCP endpoints and audit exports. Useful for inspecting a harness's host-side activity; monitoring is observational by default, while selected-action policies and MCP gateways need explicit setup." aiboost_harness,394,Design Primitives > Observability & Tracing,OpenLLMetry,https://github.com/traceloop/openllmetry,,,,,,repository,"OpenTelemetry-based instrumentation for LLM calls and agent steps: adds trace spans to every inference and tool call without modifying business logic. The cleanest way to bring the existing OTEL ecosystem (Grafana, Datadog, Jaeger) to a harness." aiboost_harness,395,Design Primitives > Observability & Tracing,Arize Phoenix,https://github.com/Arize-ai/phoenix,,,,,,repository,"Self-hostable trace UI and eval runtime for agent workflows. Lets harness engineers audit and replay every reasoning step and tool call offline, without sending data to a third-party cloud." aiboost_harness,396,Design Primitives > Observability & Tracing,Opik,https://github.com/comet-ml/opik,,,,,,repository,"Comet's open-source AI observability and evaluation platform: deep tracing of LLM calls, conversation logging, and agent activity, plus built-in eval metrics, prompt versioning, guardrails, and the Opik Agent Optimizer. Worth including because it unifies observability, verification, and optimization in one self-hostable stack rather than stitching together separate tools." aiboost_harness,397,Design Primitives > Observability & Tracing,Langfuse,https://github.com/langfuse/langfuse,,,,,,repository,"The most widely adopted self-hostable LLM observability platform: traces every agent step, manages prompt versions, and runs evals in one tool. Preferred over cloud-only alternatives when data residency or cost control is a constraint." aiboost_harness,398,Design Primitives > Observability & Tracing,Weights & Biases Weave,https://github.com/wandb/weave,,,,,,repository,"W&B's tracing and eval layer purpose-built for agent workflows: automatic call graph capture, dataset versioning, and LLM-as-judge evals that integrate directly with the wandb experiment tracking ecosystem." aiboost_harness,399,Design Primitives > Observability & Tracing,OTel GenAI Semantic Conventions,https://opentelemetry.io/docs/specs/semconv/gen-ai/,,,,,,grey,"OpenTelemetry's standard attribute names for GenAI spans (gen_ai.system, gen_ai.request.model, etc.). The naming baseline that makes harness traces portable across any OTEL-compatible backend." aiboost_harness,400,Design Primitives > Observability & Tracing,Pydantic Logfire,https://github.com/pydantic/logfire,,,,,,repository,"AI observability platform from the Pydantic team with a unique angle: all trace data is SQL-queryable (PostgreSQL-compatible), so coding agents can query production observability data directly via the Logfire MCP server. Full-stack OTEL tracing covers both the AI layer and backend — letting you determine whether a failure is in agent logic or infrastructure. The natural observability choice for Py" aiboost_harness,401,Design Primitives > Observability & Tracing,Helicone,https://github.com/Helicone/helicone,,,,,,repository,"Open-source LLM observability proxy (YC W23) with the largest open-source pricing database (300+ models). One-line proxy integration provides cost tracking, token monitoring, session tracing, and prompt versioning across providers. The AI Gateway component handles request routing and caching with zero-code changes. SOC 2 and GDPR compliant, self-hostable via Docker — the natural complement to exec" aiboost_harness,402,Design Primitives > Observability & Tracing,OpenObserve: Unified Observability for LLM Agents,https://openobserve.ai/,,,,,,grey,"2026-standard platform for LLM tracing with infrastructure log/metric unification. Enables harness engineers to correlate agent decisions with system-level events (network delays, GPU memory pressure) that explain agent failures, going beyond isolated LLM call traces." aiboost_harness,403,Design Primitives > Observability & Tracing,Braintrust,https://www.braintrust.dev,,,,,,grey,"Evaluation-first agent observability platform ($80M Series B, Feb 2026) with exhaustive auto-tracing that captures every LLM call, tool invocation, and retrieval step as nested span hierarchies. Brainstore, its purpose-built data store, enables full-trace search without sampling — critical for debugging multi-turn agent failures where the root cause spans multiple steps. Used by Stripe, Notion, Dr" aiboost_harness,404,Design Primitives > Observability & Tracing,Building Observable AI Agents: Temporal Now Integrates with Braintrust,https://temporal.io/blog/building-observable-ai-agents-temporal-now-integrates-with-braintrust,,,,,,grey,"Combines Temporal's durable execution (automatic retries, state persistence, event history replay) with Braintrust's LLM tracing so every Workflow and Activity becomes a Braintrust span and every LLM call is traced with full context. Demonstrates the pattern with a deep research agent where failed synthesis steps retry without re-executing prior searches, and prompt updates propagate via braintrus" aiboost_harness,405,Design Primitives > Observability & Tracing,Introducing BigQuery Agent Analytics,https://cloud.google.com/blog/products/data-analytics/introducing-bigquery-agent-analytics/,,,,,,grey,"Google Cloud's 2026 launch treats agent traces, tool calls, sessions, and outcomes as analytical data rather than dashboard exhaust. The important harness idea is that observability becomes queryable infrastructure: once telemetry lands in BigQuery, teams can build evaluators, regressions, and conversational debugging directly on top of production traces instead of maintaining a separate analysis " aiboost_harness,406,Design Primitives > Observability & Tracing,Distributed Tracing for Agentic Workflows with OpenTelemetry,https://developers.redhat.com/articles/2026/04/06/distributed-tracing-agentic-workflows-opentelemetry,,,2026,4,text_date,grey,"Red Hat's April 6, 2026 guide is one of the few concrete references that walks through context propagation across routing agents, specialist agents, MCP servers, and external systems using standard tracing infrastructure. It belongs here because it treats agent observability as a distributed-systems problem, which is exactly how these harnesses fail in production." aiboost_harness,407,Design Primitives > Observability & Tracing,Red-Teaming Anthropic's Internal Agent Monitoring Systems — METR,https://metr.org/blog/2026-03-25-red-teaming-anthropic-agent-monitoring/,,,2026,3,url_date,grey,"METR's three-week adversarial audit of Anthropic's internal agent monitoring and security systems (described in the Opus 4.6 Sabotage Risk Report). Discovered several novel vulnerabilities, some since patched. The most concrete published account of what it takes to stress-test agent monitoring infrastructure — essential reading before trusting any monitoring system as a safety layer." aiboost_harness,408,Design Primitives > Observability & Tracing,Future AGI,https://github.com/future-agi/future-agi,,,,,,repository,"Open-source, self-hostable platform unifying tracing, evals, simulations, guardrails, and gateway into a single feedback loop. Worth including because it demonstrates what a unified observability-and-improvement plane looks like rather than stitching together five separate vendor tools." aiboost_harness,409,Design Primitives > Observability & Tracing,agentacct,https://github.com/mikehasa/agentacct,,,,,,repository,"Local-first Agent Work Intelligence for coding agents: ingests existing Claude Code, Codex, and OpenCode session logs, attributes tokens and estimated cost to recorded work steps with confidence labels, and surfaces the evidence on a private dashboard with no cloud sync or API keys." aiboost_harness,413,Design Primitives > Debugging & Developer Experience,AgentOps,https://github.com/AgentOps-AI/agentops,,,,,,repository,"Open-source agent engineering platform (YC W24) with session replay, cost tracking, and failure detection across 10+ frameworks including CrewAI, LangGraph, and OpenAI Agents SDK. The step-by-step execution graph and cross-session metrics make it the most practical debugging layer for multi-agent systems in production." aiboost_harness,414,Design Primitives > Debugging & Developer Experience,claude-devtools,https://github.com/matt1398/claude-devtools,,,2026,2,text_date,repository,"February 2026 open-source DevTools for Claude Code that reconstructs hidden session internals from local logs: per-turn token attribution across 7 context categories, full subagent execution trees with cost breakdowns, and syntax-highlighted diffs for every tool call. Essential because Claude Code's default UI deliberately collapses tool details and thinking steps — this tool restores the visibili" aiboost_harness,415,Design Primitives > Debugging & Developer Experience,mcpsnoop,https://github.com/kerlenton/mcpsnoop,,,,,,repository,"Wireshark for MCP: a transparent proxy that shows every JSON-RPC frame between your real client and MCP servers, live in your terminal. Worth including because MCP Inspector tests servers from its own client, so it can't see the calls your actual agent makes, misses, or hangs on — mcpsnoop sits in the real data path and makes those failures visible." aiboost_harness,416,Design Primitives > Debugging & Developer Experience,Claude Code /doctor,https://code.claude.com/docs/en/commands,,,2026,7,text_date,grey,"Anthropic's July 2026 bundled setup-checkup skill (alias /checkup) that audits harness hygiene: it deduplicates local and checked-in CLAUDE.md files, flags unused skills/MCP servers/plugins, identifies slow hooks, and proposes fixes only after confirmation. The clearest first-party example of treating harness configuration drift — not model output — as the debugging target." aiboost_harness,417,Design Primitives > Debugging & Developer Experience,Syncause/debug-skill,https://github.com/Syncause/debug-skill,,,2026,4,text_date,repository,"April 2026 agent debugging skill that stops guesswork with runtime evidence. Uses background tracing (Runtime Facts) to capture the exact execution path leading to failures, then constrains the agent to cite specific data points (stack traces, variable snapshots) before proposing fixes. Moves agent debugging from ""patch and pray"" to evidence-based repair with reviewable results." aiboost_harness,418,Design Primitives > Debugging & Developer Experience,AgentTrace: Causal Graph Tracing for Root Cause Analysis in Multi-Agent Systems,https://arxiv.org/abs/2603.14688,2603.14688,,2026,3,arxiv_id,academic,March 2026 framework that localizes root causes in multi-agent execution traces using causal graph analysis rather than LLM inference. Processes traces in 0.12 seconds (69× faster than LLM-based analysis) with 93.6–95.8% accuracy across 550 synthetic failure scenarios. Distinguishes root causes from downstream symptom propagation — the key capability missing from most trace-inspection debugging wo aiboost_harness,419,Design Primitives > Debugging & Developer Experience,TraceCoder: A Trace-Driven Multi-Agent Framework for Automated Debugging of LLM-Generated Code,https://arxiv.org/abs/2602.06875,2602.06875,,2026,2,arxiv_id,academic,"February 2026 ICSE paper introducing a collaborative multi-agent debugging loop: Instrumentation Agent injects diagnostic probes, Analysis Agent performs causal trace diagnosis with a Historical Lesson Learning Mechanism (HLLM) that distills insights from prior failures, and Repair Agent executes validated fixes with rollback on regression. Achieves up to 34.43% relative improvement in Pass@1 over" aiboost_harness,420,Design Primitives > Debugging & Developer Experience,AgentRx: Systematic Debugging for AI Agents,https://www.microsoft.com/en-us/research/blog/systematic-debugging-for-ai-agents-introducing-the-agentrx-framework/,,,,,,grey,"Framework for automated root-cause analysis of agent failures: trajectory normalization, constraint synthesis from tool schemas, and constraint-guided evaluation. Achieves 23.6% better failure localization than existing approaches with a 115-trajectory annotated benchmark. Shifts agent debugging from manual log inspection to systematic constraint-based diagnosis — a reference design for harness-le" aiboost_harness,421,Design Primitives > Debugging & Developer Experience,Debugging Deep Agents with LangSmith,https://blog.langchain.com/debugging-deep-agents-with-langsmith/,,,,,,grey,"Addresses the core problem of debugging agents that run for minutes, span hundreds of steps, and produce massive traces no human can manually scan. Introduces Polly (an AI assistant that analyzes traces to surface root causes) and langsmith-fetch (CLI for piping trace data to coding agents). Key insight: debugging deep agents requires AI-assisted trace analysis — the volume of data these systems p" aiboost_harness,422,Design Primitives > Debugging & Developer Experience,Where LLM Agents Fail and How They Can Learn From Failures (AgentDebug),https://arxiv.org/abs/2509.25370,2509.25370,,2025,9,arxiv_id,academic,"](https://arxiv.org/abs/2509.25370) — ICLR 2026 paper introducing the Agent Error Taxonomy — a modular classification covering memory, reflection, planning, action, and system-level failures. The AgentDebug framework isolates root-cause failures and provides corrective feedback, achieving +24% higher all-correct accuracy. The Agent Error Benchmark (annotated trajectories from ALFWorld, GAIA, WebSh" aiboost_harness,423,Design Primitives > Debugging & Developer Experience,AgentPrism,https://github.com/evilmartians/agent-prism,,,,,,repository,"Open-source React component library (Evil Martians) that transforms OpenTelemetry trace data into interactive visualizations: tree view, timeline/Gantt view, sequence diagrams, and detail panels. Framework-agnostic — works with any OTEL-compatible agent. Fills the gap between raw OTEL spans and human-comprehensible agent debugging UIs." aiboost_harness,424,Design Primitives > Debugging & Developer Experience,Characterizing Faults in Agentic AI,https://arxiv.org/abs/2603.06847,2603.06847,,2026,3,arxiv_id,academic,"March 2026 empirical study mining 375 GitHub issues across real-world agent systems (AutoGen, CrewAI, OpenAI Agents SDK, LangChain, CAMEL, DB-GPT) to build the first grounded taxonomy of agent-specific faults: initialization failures, role deviation, memory/state deficiencies, orchestration failures, and tool integration errors. Provides architecture-level fault classification that harness enginee" aiboost_harness,425,Design Primitives > Debugging & Developer Experience,More Visibility into Copilot Coding Agent Sessions,https://github.blog/changelog/2026-03-19-more-visibility-into-copilot-coding-agent-sessions/,,,2026,3,text_date,grey,"GitHub's March 19, 2026 changelog is short but materially useful: setup-step logs, collapsed subagent traces, and clearer session-stage visibility are exactly the kind of DX improvements that make long-running agent failures debuggable in practice. It is a concrete reminder that trace readability is part of the harness, not an afterthought layered on top." aiboost_harness,426,Design Primitives > Debugging & Developer Experience,AgentStepper: Interactive Debugging of Software Development Agents,https://arxiv.org/abs/2602.06593,2602.06593,,2026,2,arxiv_id,academic,"February 2026 interactive debugger for agent execution trajectories that organizes raw logs into structured, side-by-side conversations (agent↔LLM and agent↔tools). Enables step-through execution, breakpoint manipulation, and mid-trajectory inspection. Developer study shows frustration scores drop from 5.4 to 2.4 (NASA TLX) and comprehension accuracy improves significantly — the first concrete evi" aiboost_harness,427,Design Primitives > Debugging & Developer Experience,mindwalk,https://github.com/cosmtrek/mindwalk,,,,,,repository,"Replays Claude Code and Codex sessions on a 3D map of your codebase, turning raw JSONL logs into a spatial view of where the agent searched, read, and edited. The glow-based visualization makes exploration drift and context pressure immediately visible — a concrete debugging primitive for reviewing whether an agent's footprint matched the intended task scope." aiboost_harness,428,Design Primitives > Debugging & Developer Experience,Better Harness,https://github.com/QoderAI/better-harness,,,,,,repository,"Evaluates coding-agent workflows across five dimensions and turns project and session evidence into prioritized, evidence-bounded findings with scoped repair actions. Runs inside Claude Code, Codex, Cursor, GitHub Copilot, and other hosts, making it a practical cross-tool diagnostic layer for improving the harness rather than just reviewing the final diff." aiboost_harness,429,Design Primitives > Debugging & Developer Experience,Why AI Coding Agents Actually Fail (It's Not the Model),https://www.faros.ai/blog/why-do-ai-coding-agents-fail,,,2026,8,text_date,grey,"](https://www.faros.ai/blog/why-do-ai-coding-agents-fail) — August 2026 empirical study grading six coding agents across 100 SWE-bench Pro tasks into ~4,000 failures with explanatory notes. The two largest failure clusters — instruction conflict (agents over-literally obeying ""DO NOT MODIFY: Tests"" boilerplate and refusing required work) and change hygiene (debug scripts and unrelated edits swept " aiboost_harness,430,Design Primitives > Debugging & Developer Experience,From Failed Trajectories to Reliable LLM Agents: Diagnosing and Repairing Harness Flaws,https://arxiv.org/abs/2606.06324,2606.06324,,2026,6,arxiv_id,academic,"Introduces HarnessFix, which compiles failed agent traces into a Harness-aware Trace Intermediate Representation (HTIR) that preserves step-level data- and control-flow, attributes failures to specific harness artifacts rather than the model, consolidates recurring diagnoses into repair-oriented flaw records, and applies scoped repair operators under regression-aware validation. The 6.3–18.4% gain" aiboost_harness,434,Design Primitives > Human-in-the-Loop,aws-samples/sample-human-in-the-loop-patterns,https://github.com/aws-samples/sample-human-in-the-loop-patterns,,,2026,3,text_date,repository,"March 2026 AWS reference implementation demonstrating four distinct HITL patterns for sensitive agent tool calls: Hook System (centralized blanket policy), Tool Context (per-tool fine-grained), Step Functions (async third-party approval via SNS), and MCP Elicitation (protocol-native real-time interactive approval). The most concrete production guide for choosing the right HITL architecture based o" aiboost_harness,435,Design Primitives > Human-in-the-Loop,Dify Human-in-the-Loop Node,https://github.com/langgenius/dify/discussions/32245,,,2026,2,text_date,repository,"February 2026 release making human oversight a native workflow primitive: suspend execution at critical decision points, expose review-and-edit UI mid-flow, and route subsequent execution based on human action (approve/reject/escalate). Demonstrates how HITL transitions from bolt-on approval gates to first-class execution-graph nodes with stateful pause/resume backed by Celery workers and Redis Pu" aiboost_harness,436,Design Primitives > Human-in-the-Loop,HITL Protocol,https://github.com/rotorstar/hitl-protocol,,,2026,2,text_date,repository,"Open standard (v0.8, February 2026) for human decisions in agent workflows: HTTP 202 + review URL pattern connecting services, agents, and humans across any messaging channel. No SDK required — ~15 lines of code for agents, reference implementations in Express/Hono/Next.js/FastAPI for services, and 13 end-to-end flows including escalation and hybrid approval." aiboost_harness,437,Design Primitives > Human-in-the-Loop,LangGraph — Human-in-the-Loop Concepts,https://langchain-ai.github.io/langgraph/concepts/human_in_the_loop/,,,,,,grey,"Systematic treatment of interrupt, breakpoint, and approve patterns: how to pause an agent mid-loop, persist state, and resume after human review. Directly addresses the harness engineering challenge of inserting human gates into long-running workflows." aiboost_harness,438,Design Primitives > Human-in-the-Loop,OpenAI Agents SDK — Human-in-the-Loop,https://openai.github.io/openai-agents-python/human_in_the_loop/,,,,,,grey,"The SDK treats pending approvals as first-class run interruptions: tools declare needs_approval (statically or via a callable policy), the paused run serializes to a storable RunState, and execution resumes after the decision — including across crashes and for approvals that outlive the process. The run-wide approval surface (tools inside nested agents surface approvals through the same outer run)" aiboost_harness,439,Design Primitives > Human-in-the-Loop,AutoGen — Human-in-the-Loop,https://microsoft.github.io/autogen/0.2/docs/tutorial/human-in-the-loop/,,,,,,grey,Explains human_input_mode (NEVER / TERMINATE / ALWAYS) and the UserProxyAgent as an approval gate. The most concrete implementation reference for adding human review nodes to a multi-agent conversation harness. aiboost_harness,440,Design Primitives > Human-in-the-Loop,Claude Agent SDK — Handle Approvals and User Input,https://platform.claude.com/docs/en/agent-sdk/user-input,,,,,,grey,"The most complete implementation reference for HITL mechanics: canUseTool callback pauses execution at every tool request with allow/deny/approve-with-changes/suggest-alternative response shapes; AskUserQuestion surfaces structured clarifications mid-task; streaming input enables mid-execution redirects. The ""approve with changes"" pattern — modifying tool input before execution — is the reference " aiboost_harness,441,Design Primitives > Human-in-the-Loop,HiL-Bench: Do Agents Know When to Ask for Help?,https://arxiv.org/abs/2604.09408,2604.09408,,2026,4,arxiv_id,academic,April 2026 benchmark that transforms well-specified tasks into judgment challenges by injecting 3–5 realistic blockers (missing critical information) and giving agents an ask_human() tool. Agents from top models achieve ~90% pass@3 with full information but performance drops significantly when blockers are present — the first systematic measure of when agents should escalate to humans rather than aiboost_harness,442,Design Primitives > Human-in-the-Loop,Human Judgment in the Agent Improvement Loop,https://blog.langchain.com/human-judgment-in-the-agent-improvement-loop/,,,2026,4,text_date,grey,"LangChain's April 9, 2026 guide closes an important gap that most HITL write-ups skip: human input is not just an approval gate at execution time, it's also supervision for improving prompts, tools, memory, and evaluators over time. Useful because it treats expert review as a structured data source for harness evolution rather than a one-off manual checkpoint." aiboost_harness,443,Design Primitives > Human-in-the-Loop,Humans and Agents in Software Engineering Loops,https://martinfowler.com/articles/exploring-gen-ai/humans-and-agents.html,,,,,,grey,"Martin Fowler defines three human-involvement postures — humans outside, in, or on the agent loop — and argues that ""humans on the loop"" (maintaining the harness rather than reviewing individual outputs) is the only approach that scales with agent throughput. The ""agentic flywheel"" section — where agents are directed to evaluate results and recommend harness improvements — is the clearest articula" aiboost_harness,444,Design Primitives > Human-in-the-Loop,Measuring AI Agent Autonomy in Practice,https://www.anthropic.com/news/measuring-agent-autonomy,,,2026,2,text_date,grey,"Anthropic's February 2026 empirical study of millions of real-world Claude Code interactions. Key finding: experienced users shift from per-action approval (20% auto-approve when new) to intervention-only oversight (40% auto-approve at 750+ sessions), and agent-initiated clarification stops grow faster than human interruptions as task complexity increases. The most data-grounded reference for desi" aiboost_harness,445,Design Primitives > Human-in-the-Loop,AutoResearchClaw HITL Co-Pilot,https://github.com/aiming-lab/AutoResearchClaw,,,2026,4,text_date,repository,"April 2026 open-source human-in-the-loop system with six intervention modes (full-auto, gate-only, checkpoint, step-by-step, co-pilot, custom), SmartPause confidence-driven dynamic suspension, and Intervention Learning from human corrections. The cost-guardrail system — aborting runs that exceed budget thresholds — makes it a practical reference for production HITL where human time is as constrain" aiboost_harness,446,Design Primitives > Human-in-the-Loop,agent-chief,https://github.com/SmileLikeYe/agent-chief,,,,,,repository,"Local-first attention orchestration layer that sits between you and every agent, alert, and feed: a three-stage worthiness engine decides whether to interrupt the human, dispatch work to an agent, or curate to memory. Worth including because it treats human attention as a scarce harness resource and turns the flood of agent-generated notifications into a structured, reviewable HITL decision rather" aiboost_harness,447,Design Primitives > Human-in-the-Loop,Happier,https://github.com/happier-dev/happier,,,,,,repository,"Open-source, end-to-end encrypted app that moves the human gate to your phone: tool approvals and permission prompts from Claude Code, Codex, OpenCode, Gemini CLI and other coding agents running on your own machines can be answered from desktop, phone or web. A practical reference for remote, cross-device approval of agent actions." aiboost_harness,448,Design Primitives > Human-in-the-Loop,Muxflow,https://github.com/gal064/muxflow,,,,,,repository,"Remote-first tmux IDE for coding agents. Agent notifications on desktop and phone plus voice replies let a human answer a waiting agent away from the machine, while the session keeps running in tmux over SSH." aiboost_harness,458,Reference Implementations > Tutorials & Educational,Learn Harness Engineering,https://walkinglabs.github.io/learn-harness-engineering/en/,,,,,,grey,"A project-based course on designing the environments, state, verification, and control systems that make Codex and Claude Code reliable. The most approachable public curriculum for learning harness engineering from first principles — each module builds a working artifact rather than summarizing concepts." aiboost_harness,459,Reference Implementations > Tutorials & Educational,御舆:解码 Agent Harness (Decoding the Agent Harness),https://github.com/lintsinghua/claude-code-book,,,,,,repository,"](https://github.com/lintsinghua/claude-code-book) — A book-length architectural teardown of Claude Code's harness: 15 chapters and 139 diagrams covering the tool system, four-stage permission pipeline, context compaction, memory, hooks, subagent scheduling, MCP integration, skills, streaming architecture, and plan mode. Every chapter explains *why* each design decision was made rather than how to" aiboost_harness,460,Reference Implementations > Tutorials & Educational,Harness Books,https://github.com/wquguru/harness-books,,,,,,repository,"Two open-source books that use Claude Code and Codex as observation targets to explain how constraint structures organize execution in real engineering environments. The clearest book-length treatment of why prompts, tools, permissions, recovery paths, and team rules form a single control plane rather than accessories around the model." aiboost_harness,461,Reference Implementations > Tutorials & Educational,ML6 x AISO Agent Workshop,https://github.com/ml6team/AISO-workshop,,,2026,2,text_date,repository,"February 2026 hands-on workshop building an AI agent from scratch with Google's Agent Development Kit (ADK) in 3 hours. Five milestones with a built-in benchmark that tracks progress from ~19% (base agent) to ~81% (with web search, PDF reader, and calculator tools). The clearest public tutorial for understanding how tool access directly translates to capability gains." aiboost_harness,462,Reference Implementations > Tutorials & Educational,mastra-ai/workshop-mastracode,https://github.com/mastra-ai/workshop-mastracode,,,2026,2,text_date,repository,"February 2026 workshop by Mastra's founders dissecting every layer of an open-source AI coding agent: stateful/resumable harness, dynamic prompt composition, workspace sandboxing, memory compaction, HITL steering, event protocols, and cost tracking. The 11-topic curriculum is the most complete public walkthrough of production coding-agent harness internals." aiboost_harness,463,Reference Implementations > Tutorials & Educational,Building Governed AI Agents,https://developers.openai.com/cookbook/examples/partners/agentic_governance_guide/agentic_governance_cookbook,,,2026,2,text_date,grey,"OpenAI's February 2026 cookbook building a complete multi-agent governance system from scratch: policy-as-code guardrails, OpenAI Traces for full observability, eval-driven design, and a distributable governance package. The most concrete first-party tutorial for making governance part of core infrastructure from day one." aiboost_harness,464,Reference Implementations > Tutorials & Educational,anthropics/claude-cookbooks,https://github.com/anthropics/claude-cookbooks,,,,,,repository,"Anthropic's official notebook collection covering orchestrator-worker patterns, parallel tool calling, programmatic tool calling (PTC), context compaction, and Agent SDK examples. The patterns/agents/ directory is the reference implementation of every orchestration pattern described in *Building Effective Agents*." aiboost_harness,465,Reference Implementations > Tutorials & Educational,huggingface/smolagents,https://github.com/huggingface/smolagents,,,,,,repository,"HuggingFace's deliberately minimal agent library (~1,000 lines of core code): the entire harness — tool validation, memory, monitoring, sandbox isolation (E2B, Docker, Pyodide) — is readable in an afternoon. The code-agent pattern (model writes Python that calls tools, eliminating JSON round-trips) is a concrete alternative loop design worth understanding." aiboost_harness,466,Reference Implementations > Tutorials & Educational,rasbt/mini-coding-agent,https://github.com/rasbt/mini-coding-agent,,,,,,repository,"Pure-Python coding agent harness (standard library only) that implements the six core harness components—live repo context, structured tools with permissions, context reduction, transcript resumption, and bounded subagents—in a single readable file. The clearest starting point for understanding how a coding agent loop actually works under the hood." aiboost_harness,467,Reference Implementations > Tutorials & Educational,shareAI-lab/learn-claude-code,https://github.com/shareAI-lab/learn-claude-code,,,,,,repository,"Step-by-step deconstruction of Claude Code as an agent harness (s01–s12). Best resource for understanding how agent loop, tool use, skills, context compaction, and task management compose in practice." aiboost_harness,468,Reference Implementations > Tutorials & Educational,AutoJunjie/awesome-agent-harness,https://github.com/AutoJunjie/awesome-agent-harness,,,,,,repository,"Curated list organized into Full Lifecycle Platforms, Task Runners, Agent Runtimes, Coding Agents. Close to this list's scope; good complementary reference." aiboost_harness,469,Reference Implementations > Tutorials & Educational,Skill Issue: Harness Engineering for Coding Agents,https://www.humanlayer.dev/blog/skill-issue-harness-engineering-for-coding-agents,,,,,,grey,"Practitioners' guide covering all harness configuration points for coding agents: system prompts, MCP tool selection, skills for progressive disclosure, sub-agents as context firewalls, hooks for deterministic control, and back-pressure verification. The central argument — that most agent failures are configuration problems, not model limitations — and the heuristic to minimize tool exposure (too " aiboost_harness,470,Reference Implementations > Tutorials & Educational,How to orchestrate agents using mission control,https://github.blog/ai-and-ml/github-copilot/how-to-orchestrate-agents-using-mission-control/,,,2025,12,text_date,grey,"GitHub's December 2025 practical guide on coordinating multiple coding agents with mission control: parallel vs. sequential execution, when to intervene, and how to review agent work productively. Shows the shift from single-agent prompts to multi-agent choreography and the harness decisions required to keep parallel agents from interfering with each other." aiboost_harness,471,Reference Implementations > Tutorials & Educational,awslabs/agentcore-samples,https://github.com/awslabs/agentcore-samples,,,,,,repository,"AWS's official sample repo is one of the most complete public walkthroughs of what ""productionizing"" an agent platform actually means: runtime, gateway, memory, identity, observability, IaC, and blueprint apps all live in one place. Worth including because it covers the harness infrastructure layer most sample repos skip and does so across multiple frameworks rather than baking in a single orchest" aiboost_harness,472,Reference Implementations > Tutorials & Educational,Engineering Trustworthy Multi-Agent Systems,https://www.ieeesmc.org/cai-2026/tutorial-3-engineering-trustworthy-multi-agent-systems/,,,2025,12,text_date,grey,"IEEE CAI 2026 tutorial (December 2025) providing a research-based practical guide for designing enterprise-ready multi-agent systems. Covers agentic patterns (ReACT, Reflection, CoT), emerging protocols (MCP, A2A), multi-layer memory structures, observability and online/offline evaluation techniques, and trustworthy AI guardrails. The most comprehensive conference tutorial on production multi-agen" aiboost_harness,473,Reference Implementations > Tutorials & Educational,agents-best-practices,https://github.com/DenisSergeevitch/agents-best-practices,,,,,,repository,"Provider-neutral Agent Skill for designing, auditing, and refactoring agentic harnesses across domains. Distills the full runtime discipline — loop budgets, typed tools, permission gates, compaction-aware memory, prompt-caching layout, and launch checklists — into an interactive reference that works for Codex, Claude Code, and any agent needing rigorous scaffolding." aiboost_harness,477,Reference Implementations > Generators & Meta-Harnesses,everything-claude-code,https://github.com/affaan-m/everything-claude-code,,,,,,repository,"Anthropic Hackathon Winner (140K+ stars). The agent harness performance optimization system: skills, instincts, memory optimization, continuous learning, security scanning, and research-first development. Production-ready agents, skills, hooks, rules, and MCP configurations evolved over 10+ months of intensive daily use building real products. Works across Claude Code, Codex, Cursor, OpenCode, and" aiboost_harness,479,Reference Implementations > Generators & Meta-Harnesses,Claude Agent SDK,https://platform.claude.com/docs/en/agent-sdk/overview,,,,,,grey,"Anthropic's official SDK that exposes Claude Code's entire harness as a programmable API: built-in tool execution loop, PreToolUse/PostToolUse hooks for interception, subagent definitions, allowedTools permission control, and session resumption. The highest-leverage starting point for building a production harness — you inherit the entire tool execution layer rather than implementing it." aiboost_harness,480,Reference Implementations > Generators & Meta-Harnesses,ruvnet/metaharness,https://github.com/ruvnet/metaharness,,,,,,repository,"Scaffold factory that turns any repo into a branded agent harness with its own npx CLI, MCP server, scoped memory, governance policy, and Darwin Mode self-evolution. The clearest open-source embodiment of ""the model is replaceable, the harness is the product"" — it generates the owned scaffolding rather than prescribing a fixed framework." aiboost_harness,481,Reference Implementations > Generators & Meta-Harnesses,revfactory/harness,https://github.com/revfactory/harness,,,,,,repository,"A meta-skill that generates domain-specific agent teams and the skills they use. Good example of harness-as-code, where the harness itself is produced by an agent." aiboost_harness,482,Reference Implementations > Generators & Meta-Harnesses,raphaelchristi/harness-evolver,https://github.com/raphaelchristi/harness-evolver,,,2026,3,text_date,repository,"March 2026 Claude Code plugin that autonomously evolves LLM agent harnesses using multi-agent proposers in isolated git worktrees, LangSmith-backed evaluation, and regression guards. Iterates on prompts, routing, retrieval, and orchestration code based on full-trace counterfactual diagnosis. The most practical published implementation of the Meta-Harness outer-loop optimization paradigm." aiboost_harness,483,Reference Implementations > Generators & Meta-Harnesses,neosigmaai/auto-harness,https://github.com/neosigmaai/auto-harness,,,2026,4,text_date,repository,"April 2026 open-source self-improving agentic system: bring your own coding agent, automatically mine failures from benchmark runs, optimize the harness through iterative edits, and gate changes against regressions. Supports Terminal-Bench 2.0 and tau-bench with Harbor and Docker evaluation backends. The PROGRAM.md pattern — human writes the optimization directive, agent executes the harness engin" aiboost_harness,484,Reference Implementations > Generators & Meta-Harnesses,agentic-harness-engineering,https://github.com/china-qijizhifeng/agentic-harness-engineering,,,,,,repository,"Observability-driven automatic evolution of coding-agent harnesses that decomposes the scaffold into seven orthogonal, git-tracked components and iteratively improves them through trace distillation and evidence-backed edits. Ranked #3 on Terminal-Bench 2.0 (84.7%) with demonstrated cross-model transfer, showing that evolved harness components generalize beyond benchmark-specific tuning." aiboost_harness,485,Reference Implementations > Generators & Meta-Harnesses,Meta-Harness: End-to-End Optimization of Model Harnesses,https://arxiv.org/abs/2603.28052,2603.28052,,2026,3,arxiv_id,academic,"Treats the entire harness (system prompt, tool definitions, context management, completion logic) as a joint optimization target rather than hand-tuning each piece. The key insight: give the proposer agent filesystem access to all prior harness candidates, scores, and execution traces — 10M-token diagnostic context vs. the 26K in prior work — so it can trace failures back to specific harness decis" aiboost_harness,486,Reference Implementations > Generators & Meta-Harnesses,Self-Harness: Harnesses That Improve Themselves,https://arxiv.org/abs/2606.09498,2606.09498,,2026,6,arxiv_id,academic,"June 2026 proposal for a self-improving harness loop: the agent mines its own failure traces, proposes minimal harness edits, and validates them through regression testing. Improves Terminal-Bench-2.0 pass rates by 20+ percentage points across MiniMax, Qwen, and GLM without requiring stronger external agents, showing that model-specific harness tuning can be automated rather than hand-engineered." aiboost_harness,487,Reference Implementations > Generators & Meta-Harnesses,HyperAgents: Self-Improving AI Systems,https://web.archive.org/web/20260408090850/https://pooya.blog/blog/hyperagents-self-improving-ai-meta-research-2026/,,,,,,grey,"Meta's framework integrating task-solving and meta-level improvement into a unified, editable program with metacognitive self-modification. Improved paper-review tasks from 0.0 to 0.710, transferred to Olympiad math grading at 0.630 improvement@50 score. Shows how agents can be designed to modify their own harness (prompts, tools, strategy) based on execution history — the ultimate meta-harness wh" aiboost_harness,488,Reference Implementations > Generators & Meta-Harnesses,AutoAgent,https://github.com/kevinrgu/autoagent,,,2026,4,text_date,repository,"Open-source library (April 2026) that automates the harness engineering loop itself: give it a task and a benchmark, and it iterates overnight on system prompts, tool configurations, agent orchestration, and routing — keeping or discarding each change based on score. In a 24-hour run, hit #1 on SpreadsheetBench (96.5%) and the top GPT-5 score on TerminalBench (55.1%), beating every hand-engineered" aiboost_harness,489,Reference Implementations > Generators & Meta-Harnesses,metaharness,https://github.com/SuperagenticAI/metaharness,,,2026,4,text_date,repository,"Open-source Python library (April 2026) that implements an outer optimization loop around executable harnesses for coding agents. Inspired by the Meta-Harness paper, it treats AGENTS.md, setup scripts, validation logic, and test flows as optimizable artifacts rather than static configs — with filesystem-backed run stores, environment snapshots, and scoped write enforcement. The most practical refe" aiboost_harness,490,Reference Implementations > Generators & Meta-Harnesses,meta-agent,https://github.com/canvas-org/meta-agent,,,2026,4,text_date,repository,"Lightweight continual harness optimizer (April 2026) built on the Claude Agent SDK. Runs an outer loop that reads task traces, rewrites harness configs, and re-evaluates — achieving 67% → 87% on tau-bench with no labeled training data. Demonstrates that even small, focused meta-harness loops can yield large reliability gains when harness configs are treated as learnable parameters." aiboost_harness,491,Reference Implementations > Generators & Meta-Harnesses,stanford-iris-lab/meta-harness,https://github.com/stanford-iris-lab/meta-harness,,,2026,4,text_date,repository,"April 2026 official implementation of the Meta-Harness paper from Stanford's IRIS Lab. Provides the framework and two reference experiments for end-to-end harness optimization via filesystem-backed search loops where a coding agent proposes, evaluates, and refines harness artifacts. The cleaned-up codebase is the definitive starting point for researchers reproducing or extending meta-harness optim" aiboost_harness,492,Reference Implementations > Generators & Meta-Harnesses,autocontext,https://github.com/greyhaven-ai/autocontext,,,,,,repository,"Recursive self-improving harness that runs multi-generation evaluation loops, distilling successful strategies into persistent playbooks and trace datasets that future agents inherit. The five-role architecture (competitor, analyst, coach, architect, curator) and built-in production trace capture make it the most complete open-source reference for turning meta-harness optimization into a deployabl" aiboost_harness,493,Reference Implementations > Generators & Meta-Harnesses,retro-harness,https://github.com/wbopan/retro-harness,,,,,,repository,"Official implementation of RHO (Retrospective Harness Optimization): improves an agent's harness using only its own past trajectories, via self-validation, self-consistency, and pairwise self-preference — no ground-truth labels or external evaluators required. A single round moves SWE-Bench Pro from 59% to 78%, showing that harness evolution can be fully self-supervised rather than dependent on la" aiboost_harness,494,Reference Implementations > Generators & Meta-Harnesses,Continual Harness,https://github.com/sethkarten/continual-harness,,,,,,repository,"Reset-free framework that lets an LLM Refiner rewrite its own system prompt, sub-agents, skills, and memory mid-episode via an evolve_harness tool. The first open-source implementation of online harness self-improvement with reproducible long-horizon benchmarks (Gemini Plays Pokémon)." aiboost_harness,495,Reference Implementations > Generators & Meta-Harnesses,Omnigent,https://github.com/omnigent-ai/omnigent,,,2026,6,text_date,repository,"Databricks' open-source meta-harness (June 2026) that sits above Claude Code, Codex, Pi, and custom agents to compose, govern, and share live agent sessions from one control plane. The key harness insight is that enterprises don't need another agent framework — they need a portability and governance layer above the agents they already run, with policy-driven sandboxing and cost caps enforced outsi" aiboost_harness,496,Reference Implementations > Generators & Meta-Harnesses,oh-my-claudecode,https://github.com/Yeachan-Heo/oh-my-claudecode,,,,,,repository,"Claude Code plugin that externalizes multi-agent orchestration as installable skills and staged team pipelines, with cross-provider advisor routing and built-in requirement-clarification interviews. The most widely adopted example of turning a single-agent CLI into a team-ready meta-harness without rewriting the underlying loop." aiboost_harness,498,Reference Implementations > Generators & Meta-Harnesses,Nexent,https://github.com/ModelEngine-Group/nexent,,,,,,repository,"Zero-code platform for auto-generating production-grade AI agents using Harness Engineering principles: unified tools, skills, memory, and orchestration with built-in constraints, feedback loops, and control planes. The clearest open-source example of turning harness scaffolding from a hand-rolled craft into a declarative, production-ready system." aiboost_harness,499,Reference Implementations > Generators & Meta-Harnesses,Beagle,https://github.com/SalesforceAIResearch/Beagle,2608.07545,,2026,8,arxiv_id,repository,"Salesforce AI Research's platform for evaluating and evolving agent harnesses at scale (September 2026), shipping the official implementation of [DarwinX](https://arxiv.org/abs/2608.07545): natural-selection search over harness changes where each trial is driven faithfully by the benchmark's own framework (Harbor, PIER) and verifier-backed tasks guard against regressions. The evolved harness match" aiboost_harness,500,Reference Implementations > Generators & Meta-Harnesses,Harness-R1: Learning to Edit Executable Runtime Harnesses from Agent Failure Trajectories,https://arxiv.org/abs/2608.02276,2608.02276,,2026,8,arxiv_id,academic,"August 2026 method that makes harness editing a *learned capability* rather than a fixed editor agent: a dedicated 9B harness engineer is post-trained with online RL — batches of target-agent failures become validated executable patches, fresh reruns of the frozen target supply outcome rewards, and only the engineer is updated. Raising vanilla Qwen3.5-9B from 44.3% to 53.6% across WebShop, ALFWorl" aiboost_harness,501,Reference Implementations > Generators & Meta-Harnesses,MemoHarness: Agent Harnesses That Learn from Experience,https://arxiv.org/abs/2607.14159,2607.14159,,2026,7,arxiv_id,academic,"July 2026 framework that decomposes the harness into six control dimensions along the inference flow (context assembly, tool interaction, generation control, orchestration, memory management, output processing) and adapts the configuration per case by retrieving from a dual-layer experience bank of per-case failure diagnoses and distilled global patterns — no labels, feedback, or search rounds at " aiboost_harness,502,Reference Implementations > Generators & Meta-Harnesses,Harness-Zero: Harness Distillation via Agent-as-Harness,https://arxiv.org/abs/2609.24974,2609.24974,,2026,9,arxiv_id,academic,"September 2026 method (Peking University, Google, HKUST) for making harness gains portable: a harnessing agent guided by an evolved reference harness reviews and corrects the student's *own* trajectories in the target harness's action space before execution, turning harness guidance into SFT demonstrations. The critical findings — the distilled model under a minimal harness beats the base model wi" aiboost_harness,503,Reference Implementations > Generators & Meta-Harnesses,Harness Evolution Hits a Ceiling: When Weight Training Should Begin,https://arxiv.org/abs/2610.11655,2610.11655,,2026,10,arxiv_id,academic,"October 2026 study of self-evolving harness loops that pinpoints where they stop working: evolution reliably repairs *process* failures (blocked calls, loops, exhausted step budgets) but stalls on *content* failures (poor-quality delivered plans), because the loop's evidence package never captures failure composition. The proposed fix is diagnose-then-intervene — keep evolving the harness for proc" aiboost_harness,504,Reference Implementations > Generators & Meta-Harnesses,"SBCO: Self-Supervised, Verifier-Grounded Harness Optimization for Planning Agents",https://arxiv.org/abs/2608.10157,2608.10157,,2026,8,arxiv_id,academic,August 2026 (Samsung Research America) method for domains where self-referential harness improvement breaks down because task competence doesn't imply self-modification competence (unlike coding). SBCO jointly learns a decomposed bank of per-constraint verifiers and a harness policy via block coordinate ascent — a fixed meta-agent grades its own outputs in a closed loop with no human labels — matc aiboost_harness,505,Reference Implementations > Generators & Meta-Harnesses,Causal Improvement Graph for Agentic Harness Optimization,https://arxiv.org/abs/2610.05039,2610.05039,,2026,10,arxiv_id,academic,"October 2026 meta-harness framework that externalizes the improvement state of harness optimization into a persistent graph: Evidence, Hypothesis, Intervention, and Outcome nodes linked so each proposer builds on the *relations* among prior findings instead of re-deriving them from an ever-growing raw history. Directly targets the scaling failure of proposer-centric meta-harnesses (Meta-Harness, S" aiboost_harness,506,Reference Implementations > Generators & Meta-Harnesses,HarnessDev: Can LLMs Create and Evolve Their Own Agent Harness?,https://arxiv.org/abs/2609.01437,2609.01437,,2026,9,arxiv_id,academic,"September 2026 benchmark that treats the harness itself as the artifact under evaluation: given a minimal seed, creator LLMs must build a complete execution system from scratch, then evolve it from downstream feedback, scored on capability and execution-token cost. Across four domains and five benchmarks (2,207 instances), generated harnesses fall substantially behind human-engineered references o" aiboost_harness,507,Reference Implementations > Generators & Meta-Harnesses,"Harness Handbook: Making Evolving Agent Harnesses Readable, Navigable, and Editable",https://arxiv.org/abs/2607.13285,2607.13285,,2026,7,arxiv_id,academic,"July 2026 paper targeting the bottleneck this section's other entries skip: before a meta-loop can improve a harness, something must locate *where* a requested behavior lives in a complex, tightly coupled harness codebase. The Harness Handbook is an automatically synthesized behavior-centric index linking harness behaviors to source code, and Behavior-Guided Progressive Disclosure navigates agents" aiboost_harness,508,Reference Implementations > Generators & Meta-Harnesses,Sage,https://github.com/ZHangZHengEric/Sage,,,,,,repository,"Agent harness with a recursive self-improvement (RSI) loop that separates task execution, evaluation, and package tuning. A reference for evidence-backed harness iteration: training feedback guides edits, fixed validation selects candidates, and adoption remains an explicit user choice." aiboost_harness,509,Reference Implementations > Generators & Meta-Harnesses,Most Capable Agent System Prompt,https://github.com/fainir/most-capable-agent-system-prompt,,,,,,repository,"A copy-paste system prompt that has a coding agent (Claude Code, Codex, Cursor) build its own harness, either as a wrapper around that agent or as a native runtime on an agent SDK. It specifies file-based project state (AGENTS.md, plan.md, tasks.md, FAILURE.md), a first milestone that proves one closed loop from goal to verified result and recorded lesson, and a build order for the control plane, " aiboost_harness,513,Reference Implementations > Demo Harnesses,Anthropic Computer Use Demo,https://github.com/anthropics/anthropic-quickstarts/tree/main/computer-use-demo,,,,,,repository,"Anthropic's reference harness for the screenshot-action loop: defines the screenshot, bash, and text_editor tool interface that makes desktop/browser control work. Essential reading before building any harness where the agent's primary sensory input is a rendered screen rather than structured API responses." aiboost_harness,514,Reference Implementations > Demo Harnesses,Codex CLI,https://github.com/openai/codex,,,,,,repository,"OpenAI's official autonomous coding agent CLI — the open-source reference implementation of the Codex harness with sandboxed tool execution, multi-file editing, and a streaming agent loop. Worth studying because it is the most widely adopted terminal-native coding agent harness and exposes the same loop architecture that OpenAI documents in its harness engineering posts." aiboost_harness,515,Reference Implementations > Demo Harnesses,coleam00/your-claude-engineer,https://github.com/coleam00/your-claude-engineer,,,,,,repository,"Agent harness with Slack, GitHub, and Linear integrations. Useful reference for how real-world tool wiring works inside a harness." aiboost_harness,516,Reference Implementations > Demo Harnesses,OpenHands,https://github.com/OpenHands/OpenHands,,,,,,repository,"The most architecturally complete open-source coding agent: Runtime/Sandbox isolation, EventStream message bus, and Agent Controller are a three-layer harness design worth studying for production deployments." aiboost_harness,517,Reference Implementations > Demo Harnesses,OpenSRE,https://github.com/Tracer-Cloud/opensre,,,,,,repository,"Open-source framework for building AI SRE agents with 60+ observability and remediation tool integrations plus a synthetic incident evaluation environment. The clearest open-source reference for turning infrastructure incident response into a trainable, evaluable agent harness rather than a one-off chatbot." aiboost_harness,518,Reference Implementations > Demo Harnesses,Goose,https://github.com/aaif-goose/goose,,,2026,4,text_date,repository,"Block's open-source, extensible AI agent donated to the Linux Foundation's Agentic AI Foundation in April 2026. Its MCP-native architecture treats every capability as an MCP server, making it a practical reference for building vendor-neutral, extensible harnesses where tool integration is the primary extension mechanism rather than framework-specific plugins." aiboost_harness,519,Reference Implementations > Demo Harnesses,browser-use,https://github.com/browser-use/browser-use,,,,,,repository,"Open-source browser agent (Python and TypeScript) whose loop injects serialized DOM state each step, exposes browser actions as registered tools, and runs the same code against a local Chrome or a cloud browser. It has grown from a minimal reference into a full framework (~70k lines of Python across agent, DOM, tools, MCP, and sandbox modules) — study it for production DOM-to-context serialization" aiboost_harness,520,Reference Implementations > Demo Harnesses,browser-harness,https://github.com/browser-use/browser-harness,,,,,,repository,"Self-healing browser harness that connects an LLM directly to your real Chrome via CDP. The critical design decision: the agent itself writes missing helpers and domain skills into the harness during execution, so the scaffold improves every run rather than requiring manual updates. At ~1k lines across four core files, it is the clearest published demonstration of an editable harness that learns f" aiboost_harness,522,Reference Implementations > Demo Harnesses,SWE-agent,https://github.com/SWE-agent/SWE-agent,,,,,,repository,"Coding agent whose Agent-Computer Interface (ACI) — purpose-built file viewer, search, and editor tools with explicit state constraints and error feedback — is the reference design for adapting a tool interface to a specific task domain rather than using generic bash." aiboost_harness,523,Reference Implementations > Demo Harnesses,Aider,https://github.com/Aider-AI/aider,,,,,,repository,"AI pair-programmer harness with an Architect mode that splits planning (one LLM) from coding (another), and git-aware tooling that uses version control as the undo mechanism instead of custom state rollback. The best reference for multi-file editing tool design and planner/coder layer separation." aiboost_harness,524,Reference Implementations > Demo Harnesses,Open SWE: An Open-Source Framework for Internal Coding Agents,https://blog.langchain.com/open-swe-an-open-source-framework-for-internal-coding-agents/,,,,,,grey,"A composable coding-agent harness built on Deep Agents, synthesizing design patterns from Stripe, Ramp, and Coinbase production deployments. Key decisions: curated ~15-tool limit enforced at harness design time, one isolated sandbox (Modal/Daytona/Runloop/LangSmith) per task, AGENTS.md for injecting repo-wide conventions, and Linear/Slack task context in the system prompt. The most recent publishe" aiboost_harness,525,Reference Implementations > Demo Harnesses,Live-SWE-agent: Autonomous Software Agent with Self-Evolving Harness,https://arxiv.org/html/2511.13646v3,2511.13646,,2025,11,arxiv_id,academic,"Production harness achieving 77.4% solve rate on SWE-bench Verified through continuous harness evolution — the scaffold adapts from failure signals rather than requiring manual retuning per task class. Demonstrates the architectural pattern where the harness itself is a learnable component, not just a static container for a fixed agent." aiboost_harness,526,Reference Implementations > Demo Harnesses,Pipecat: Python Framework for Real-Time Voice Agent Pipelines,https://github.com/pipecat-ai/pipecat,,,,,,repository,"Handles frame management, streaming media coordination, and pipeline orchestration between ASR/LLM/TTS services for sub-800ms Total Turn-Around Time voice interactions. The missing harness primitive for voice agents: manages backpressure, handles frame queueing, and exposes a simple async interface for real-time constraints. Critical infrastructure for building responsive voice-first agents." aiboost_harness,527,Reference Implementations > Demo Harnesses,Qwen Audio Agent,https://github.com/QwenAudio/qwen-audio-agent,,,,,,repository,"Qwen's realtime voice runtime that keeps agents talking, working, and present across multiple backend agents (Claude Code, Codex, Qwen Code, Kimi Code). It demonstrates how to wrap terminal-native coding agents in a persistent, interruptible voice shell without losing task continuity — a concrete reference for ambient agent interfaces." aiboost_harness,528,Reference Implementations > Demo Harnesses,The Virtual Biotech: Multi-Agent AI Framework for Drug Discovery,https://www.biorxiv.org/content/10.64898/2026.02.23.707551v1,,,,,,academic,"Orchestrated team of domain-specialized scientist agents that autonomously analyzed 55,984 clinical trials and discovered cell-type-specific drug targets 40% more likely to succeed Phase I→II transitions. Demonstrates specialized harness design for scientific workflows where formal reasoning, multi-agent coordination, and domain-specific tool suites are load-bearing constraints. Shows the upper bo" aiboost_harness,529,Reference Implementations > Demo Harnesses,"Building NVIDIA Nemotron 3 Agents for Reasoning, Multimodal RAG, Voice, and Safety",https://developer.nvidia.com/blog/building-nvidia-nemotron-3-agents-for-reasoning-multimodal-rag-voice-and-safety/,,,,,,grey,"NVIDIA's Nemotron 3 family (Super for long-context reasoning, Content Safety for multimodal moderation, VoiceChat for real-time speech) designed for scalable agentic AI with enterprise-grade multimodal understanding. The integration of safety models, vision models, and voice models into a single harness stack is the reference architecture for production multimodal agents." aiboost_harness,530,Reference Implementations > Demo Harnesses,AIO Sandbox,https://github.com/agent-infra/sandbox,,,,,,repository,"All-in-one agent sandbox combining browser, shell, filesystem, MCP servers, and VSCode Server in a single Docker container. Native MCP support exposes sandbox capabilities to LLMs via the standard protocol, and files downloaded in the browser are instantly accessible in terminal and VSCode. Optimized startup (4–8s depending on config) with Claude Skills mounting support. The fastest path to a full" aiboost_harness,531,Reference Implementations > Demo Harnesses,GitHub Agentic Workflows,https://github.blog/changelog/2026-02-13-github-agentic-workflows-are-now-in-technical-preview/,,,2026,2,text_date,grey,"GitHub's February 13, 2026 technical preview is unusually valuable because the implementation is fully open source (gh-aw) and shows how natural-language workflow generation, approval handling, and GitHub-native execution fit together in one harness. It belongs here as a reference implementation for teams that want to study agentized CI/CD rather than just chat-centric coding agents." aiboost_harness,532,Reference Implementations > Demo Harnesses,langchain-ai/deepagents,https://github.com/langchain-ai/deepagents,,,2026,4,text_date,repository,"LangChain's batteries-included agent harness (released April 2026) with built-in planning, filesystem tools, shell access, sub-agents, and auto-summarization. The clearest open-source demonstration of how a general-purpose coding agent harness can be made ready-to-run out of the box while remaining fully extensible." aiboost_harness,533,Reference Implementations > Demo Harnesses,HKUDS/OpenHarness,https://github.com/HKUDS/OpenHarness,,,2026,4,text_date,repository,"A compact, inspectable open-source agent harness from HKUDS (April 2026) featuring a built-in personal agent (ohmo), auto-compaction with session preservation, MCP HTTP transport, and multimodal gateway support. Excellent reference for understanding how a small, modular harness can support multi-day sessions without manual context management." aiboost_harness,534,Reference Implementations > Demo Harnesses,nanobot,https://github.com/HKUDS/nanobot,,,,,,repository,"HKUDS's ultra-lightweight, self-hosted personal AI agent framework: a single readable Python core that combines WebUI/terminal/chat-app surfaces, long-term memory, MCP tools, model routing, multi-agent delegation, and scheduled automation. A practical reference for how a complete personal agent harness can own the full stack without becoming a black-box platform." aiboost_harness,536,Reference Implementations > Demo Harnesses,OpenCode,https://github.com/anomalyco/opencode,,,,,,repository,"Open-source terminal-native AI coding agent with 131K+ stars and 2.5M+ monthly active developers. Provider-agnostic architecture supports 75+ LLM providers plus native LSP auto-configuration, multi-session parallel agents, and MCP extensibility. The build/plan agent split and client/server architecture make it the most complete open-source reference for a terminal-first coding harness." aiboost_harness,537,Reference Implementations > Demo Harnesses,Squad,https://github.com/bradygaster/squad,,,,,,repository,"Repository-native multi-agent orchestration framework built on GitHub Copilot. Initializes a persistent AI team (lead, frontend, backend, tester) as files inside your repo — knowledge compounds across sessions through committed history.md and decisions.md. The most accessible reference for teams that want multi-agent coordination without heavy infrastructure." aiboost_harness,539,Reference Implementations > Demo Harnesses,AMAP-ML/LongHorizon-Harness,https://github.com/AMAP-ML/LongHorizon-Harness,,,2026,8,text_date,repository,"Long-horizon computer-use harness that runs on top of Claude Code and Codex, splitting work into Manager, Executor, and Auditor roles so only independently verified results enter persistent task state. The clearest August 2026 open-source reference for carrying desktop-and-CLI tasks through dozens of hours without state drift." aiboost_harness,540,Reference Implementations > Demo Harnesses,ClawGUI,https://github.com/ZJU-REAL/ClawGUI,,,,,,repository,"End-to-end harness for GUI agents that unifies online RL training, standardized benchmarking, and real-device deployment in one framework. The most complete open-source reference for building visual perception-action loops across desktop and mobile surfaces." aiboost_harness,541,Reference Implementations > Demo Harnesses,AOHP,https://github.com/aohp-os/aohp,,,,,,repository,"Android Open Harness Project: an open-source OS-level agent harness built on AOSP that treats AI agents as first-class OS actors, redesigning service composition, agent interfaces, and information flow for personalized, efficient, and secure mobile interaction. It fills a rare gap by moving the scaffolding from an application layer down into the operating system itself, with early results reportin" aiboost_harness,542,Reference Implementations > Demo Harnesses,desloppify,https://github.com/peteromallet/desloppify,,,,,,repository,"Agent harness that turns codebase quality improvement into a structured, score-driven workflow: mechanical detectors find dead code and complexity, LLM review assesses naming and abstractions, and a persistent next → fix → resolve loop keeps the agent on track across sessions. The anti-gaming score design — where wontfix items widen the gap between lenient and strict scores — is a rare example of " aiboost_harness,543,Reference Implementations > Demo Harnesses,DeepSeek-Reasonix,https://github.com/esengine/DeepSeek-Reasonix,,,,,,repository,DeepSeek-native coding agent harness engineered around prefix-cache stability as a loop invariant: immutable-prefix / append-only-log / volatile-scratch partitioning achieves 99.82% cache-hit rates and ~5× cost reduction on long sessions. The most detailed public case study of designing an entire agent loop to preserve a provider-specific economic property rather than treating caching as an aftert aiboost_harness,544,Reference Implementations > Demo Harnesses,dao-code,https://github.com/tigicion/dao-code,,,,,,repository,"DeepSeek-V4 terminal coding agent harness that treats prefix-cache economics as a first-class design constraint: byte-stable system prompts, cache-reusing forks for reflection and memory, and a self-verifying cross-session memory layer keep real SWE-bench-style tasks at ~95.8% cache hit and ~30× cheaper than Claude Opus. A concrete demonstration that harness-level cache engineering can substitute " aiboost_harness,545,Reference Implementations > Demo Harnesses,SmallCode,https://github.com/Doorman11991/smallcode,,,,,,repository,"Terminal-native coding agent built from the ground up for 8B–35B local models. Its harness innovations are all compensations for small-model limitations: 2-stage tool routing halves schema overhead, a forgiving multi-format parser recovers from malformed JSON/YAML/XML tool calls, patch-first editing avoids the truncation errors common when small models rewrite entire files, and a context budget en" aiboost_harness,546,Reference Implementations > Demo Harnesses,alibaba/open-code-review,https://github.com/alibaba/open-code-review,,,,,,repository,"Alibaba's production-grade code review agent combining deterministic pipelines with LLM reasoning: built-in fine-tuned rules catch NPE, thread-safety, and injection vulnerabilities at line-level precision, while the LLM layer handles nuanced design feedback. Demonstrates how hybrid harnesses can outperform purely model-driven or purely static-analysis approaches by assigning each layer the problem" aiboost_harness,547,Reference Implementations > Demo Harnesses,DeerFlow,https://github.com/bytedance/deer-flow,,,,,,repository,"ByteDance's open-source SuperAgent harness built on LangGraph: orchestrates sub-agents with isolated contexts, persistent multi-tier memory, Docker/K8s sandbox execution, and on-demand skill loading for long-horizon tasks that span minutes to hours. A concrete reference for composing supervisor coordination, scoped delegation, and real execution environments into a single deployable harness." aiboost_harness,549,Reference Implementations > Demo Harnesses,grok-build,https://github.com/xai-org/grok-build,,,,,,repository,"SpaceXAI's open-source terminal coding agent harness: a Rust-based fullscreen TUI with an extensible tool runtime, MCP/skills/hooks support, and headless/embedded modes via ACP. A useful first-party counterpoint to Claude Code and Codex CLI for studying how a new model provider structures the loop, checkpoints workspace state, and exposes the same harness through interactive and scripted surfaces." aiboost_harness,550,Reference Implementations > Demo Harnesses,CodeWhale,https://github.com/Hmbown/CodeWhale,,,,,,repository,"Open-source, community-driven terminal coding agent harness written in Rust with a model-agnostic runtime, OS-level sandbox (Seatbelt/Landlock/seccomp/bwrap), resumable fleets via an append-only ledger, and a /model command that lets you switch providers mid-task. A strong example of treating the harness — not the model — as the product surface, with clear separation between interactive TUI, headl" aiboost_harness,551,Reference Implementations > Demo Harnesses,FrontierAgent,https://github.com/ApodexAI/FrontierAgent,,,2026,8,text_date,repository,"Apodex's open-source agent runtime (August 2026) shipping two native workflows — ReAct and Agent Team (a coordinator maintains a live task board, delegates bounded work to parallel sub-agents, and synthesizes their reports). The design decisions are the value: a task-scoped filesystem contract (/inputs read-only, /workspace mutable, /outputs as persistent deliverables) with fail-closed authorizati" aiboost_harness,552,Reference Implementations > Demo Harnesses,DeepSeek Harness (dsh),https://github.com/deepseek-ai/deepseek-harness,,,2026,8,text_date,repository,"](https://github.com/deepseek-ai/deepseek-harness) — DeepSeek's official open-source agent harness (August 2026) built on an ""everything is a plugin"" architecture powered by Cordis: models, tools, skills, UI surfaces, and even the desktop shell are plugins composed through dependency injection rather than a monolith with extension points, with an accompanying paper on spatiotemporal composability." aiboost_harness,553,Reference Implementations > Demo Harnesses,ZCode,https://github.com/zai-org/ZCode,,,2026,9,text_date,repository,"Z.ai's official open-source coding agent harness (September 2026, Apache-2.0): a single TypeScript agent runtime — CLI, TUI, tools, and an RPC client SDK — served through three client surfaces (Electron desktop, browser, terminal) from one zcode command. The clearest recent first-party reference for the ""one agent core, many client surfaces"" split, complementing grok-build and DeepSeek Harness for" aiboost_harness,554,Reference Implementations > Demo Harnesses,Kimi Code CLI,https://github.com/MoonshotAI/kimi-code,,,2026,5,text_date,repository,"Moonshot AI's official open-source coding agent harness (May 2026): a single-binary, millisecond-startup TUI with built-in coder/explore/plan subagents in isolated contexts, lifecycle hooks, conversational MCP configuration (/mcp-config instead of hand-edited JSON), and first-class video input as a tool surface. Speaking ACP out of the box (Zed, JetBrains) makes it a useful first-party reference f" aiboost_harness,555,Reference Implementations > Demo Harnesses,unreal-agent,https://github.com/unreallabsai/unreal-agent,,,2026,9,text_date,repository,"Unreal Labs' async-first agent harness in Go (September 2026) with an unusually disciplined architecture: a single coordinator event loop, tool translators that validate synchronously and emit serializable operations for async execution, a forkable append-only session store with explicit versioned formats, and a context builder that returns the model input *together with a record of everything it " aiboost_harness,556,Reference Implementations > Demo Harnesses,DeepCode,https://github.com/HKUDS/DeepCode,,,2026,9,text_date,repository,"HKUDS's open agentic coding harness (September 2026) built around goal-driven loop engineering: the agent analyzes, implements, verifies, and repairs against a natural-language Goal while the user can revise the Goal, queue instructions, or pause mid-loop. The most instructive design decisions are evidence-driven completion — selecting task-appropriate verification (tests, builds, static checks, d" aiboost_harness,557,Reference Implementations > Demo Harnesses,Mercury CLI,https://github.com/Whq02/MercuryCLI,,,,,,repository,"Apache-2.0 terminal coding agent harness in which each session writes a handoff note when it finishes and the next session reads it once at startup alongside the project's pinned rules, while parallel crewmates work in isolated git worktrees under project file leases. Useful for seeing how state between sessions and several agents sharing one repository are handled in one harness." aiboost_harness,562,Reference Implementations > Adjacent Collections,EvoMap/awesome-agent-evolution,https://github.com/EvoMap/awesome-agent-evolution,,,2026,4,text_date,repository,"April 2026 curated list covering agent evolution, memory systems, multi-agent architectures, and self-improvement. Complements this list with a forward-looking lens on the next generation of agent capabilities — where harnesses must adapt to agents that modify their own scaffolding over time." aiboost_harness,563,Reference Implementations > Adjacent Collections,Picrew/awesome-agent-harness,https://github.com/Picrew/awesome-agent-harness,,,2026,4,text_date,repository,"Implementation-first curated list (April 2026) with 150 entries, 84% GitHub projects, organized into 9 categories from harness architecture to sandboxing. The featured blogs section and catalog-style organization make it a strong complementary reference to this list's article-centric approach." aiboost_harness,564,Reference Implementations > Adjacent Collections,jiji262/awesome-harness-engineering,https://github.com/jiji262/awesome-harness-engineering,,,,,,repository,"Focuses on platform delivery governance, IDP, GitOps, and AI-native engineering. Overlaps with this list on the platform engineering side; more Harness-the-company oriented." aiboost_harness,565,Reference Implementations > Adjacent Collections,VoltAgent/awesome-ai-agent-papers,https://github.com/VoltAgent/awesome-ai-agent-papers,,,,,,repository,"Curated collection of 363+ arXiv papers from 2026 organized into five harness-relevant categories: Multi-Agent (51), Memory & RAG (56), Eval & Observability (79), Agent Tooling (95), AI Agent Security (82). Weekly updates make it the best single source for tracking research that will shape harness design decisions in 2026." aiboost_harness,566,Reference Implementations > Adjacent Collections,bradAGI/awesome-cli-coding-agents,https://github.com/bradAGI/awesome-cli-coding-agents,,,,,,repository,"Catalog of 80+ terminal-native AI coding agents (open-source and proprietary) plus the harnesses that orchestrate, sandbox, and extend them: session managers, parallel runners, autonomous loop infrastructure, and credential vaults. The most comprehensive reference for the CLI agent layer that most harness infrastructure is designed to host." aiboost_harness,567,Reference Implementations > Adjacent Collections,danielrosehill/AI-Harnesses,https://github.com/danielrosehill/AI-Harnesses,,,2026,4,text_date,repository,"April 2026 point-in-time snapshot of projects describing themselves as AI agent harnesses, organized into Resource Lists, Harness Runtimes, and Reference Implementations. Useful as a landscape survey of how the term ""harness"" is being applied across the ecosystem — from lightweight wrappers to full orchestration frameworks." aiboost_harness,568,Reference Implementations > Adjacent Collections,RyanAlberts/best-of-Agent-Harnesses,https://github.com/RyanAlberts/best-of-Agent-Harnesses,,,,,,repository,"A curated, ranked list of 124 agent harnesses, rescored weekly and published as machine-readable data with an MCP server. The most practical complement for discovering and comparing harnesses, and a rare example of a list built to be consumed by agents themselves." aiboost_harness,574,"Security, Sandbox & Permissions",Beyond Permission Prompts,https://www.anthropic.com/engineering/claude-code-sandboxing,,,,,,grey,The authoritative resource on moving from prompt-level permission grants to structured authorization in the harness. aiboost_harness,575,"Security, Sandbox & Permissions",How we contain Claude across products,https://www.anthropic.com/engineering/how-we-contain-claude,,,2026,5,text_date,grey,"Anthropic's May 2026 cross-product containment write-up: why environmental isolation must be the primary boundary, how model-layer defenses alone miss ~17% of overeager actions, and concrete sandbox architectures for chat, terminal, and autonomous workspace products. The exfiltration-through-allowlist case study is a stark reminder that the weakest link is often custom harness plumbing, not the sa" aiboost_harness,576,"Security, Sandbox & Permissions","Building a safe, effective sandbox to enable Codex on Windows",https://openai.com/index/building-codex-windows-sandbox/,,,2026,7,text_date,grey,"OpenAI's July 2026 engineering deep-dive into sandboxing Codex on Windows, where no Seatbelt/seccomp-style capability isolation exists: why AppContainer, Windows Sandbox, and Mandatory Integrity Control fell short, and how write-restricted tokens plus network suppression produce an elevated sandbox that keeps file writes and network access inside safe bounds. The clearest first-party reference for" aiboost_harness,577,"Security, Sandbox & Permissions",The Agent Harness Belongs Outside the Sandbox,https://www.mendral.com/blog/agent-harness-belongs-outside-sandbox,,,2026,4,text_date,grey,"Andrea Luzzardi's April 2026 argument for running the agent loop outside the sandbox: credentials stay out of untrusted containers, sandboxes become suspendable cattle rather than session lifelines, and the architecture cleanly separates orchestration from execution. The clearest published case for why ""harness inside"" and ""harness outside"" are different security and reliability models, not implem" aiboost_harness,578,"Security, Sandbox & Permissions",Build zero-trust AI agents with Google's Agent Development Kit,https://developers.googleblog.com/build-zero-trust-ai-agents-with-googles-agent-development-kit/,,,2026,8,text_date,grey,"Google's August 2026 guide to hardening autonomous ADK agents against prompt injection and production-state mutation, using cryptographic write signatures, gVisor sandboxing, and deterministic semantic gateways that sit outside the LLM context. The accompanying open-source customer-support agent demonstrates the threat model and defenses concretely." aiboost_harness,579,"Security, Sandbox & Permissions",Model Context Protocol — Authorization,https://modelcontextprotocol.io/specification/2026-07-28/basic/authorization,,,2026,7,url_date,grey,MCP's specification for OAuth-based authorization flows when agents access external services. aiboost_harness,580,"Security, Sandbox & Permissions",AgentShield,https://github.com/affaan-m/agentshield,,,,,,repository,"Security scanner for AI agent configurations that detects hardcoded secrets, permission misconfigurations, hook injection, risky MCP servers, and prompt-injection vectors in Claude Code setups. Worth including because it turns harness security from a post-hoc audit into a pre-commit gate, surfacing configuration-level vulnerabilities before the agent runs." aiboost_harness,581,"Security, Sandbox & Permissions",HOL Guard,https://github.com/hashgraph-online/hol-guard,,,,,,repository,"Apache-2.0 action-policy checks and human approvals for supported AI coding agents; helps harness maintainers gate shell, file, MCP, and package operations before execution where supported." aiboost_harness,583,"Security, Sandbox & Permissions",tldrsec/prompt-injection-defenses,https://github.com/tldrsec/prompt-injection-defenses,,,,,,repository,"The most complete catalog of practical prompt injection defenses (input validation, tool output sanitization, canary tokens, etc.). Functions as a design checklist for hardening trust boundaries in any agent harness." aiboost_harness,584,"Security, Sandbox & Permissions",Prompt Injection — Simon Willison's Series,https://simonwillison.net/series/prompt-injection/,,,,,,grey,"The most thorough public writing on why indirect prompt injection is uniquely dangerous for agent harnesses: agents actively consume untrusted external content (emails, web pages, tool outputs) that can hijack their actions. Essential for understanding the attack surface before designing trust boundaries." aiboost_harness,585,"Security, Sandbox & Permissions",OWASP LLM01:2025 — Prompt Injection,https://genai.owasp.org/llmrisk/llm01-prompt-injection/,,,,,,grey,"OWASP's authoritative classification of direct and indirect prompt injection risks. Complements the tldrsec defense catalog: use this to define the threat model, use tldrsec to select countermeasures." aiboost_harness,586,"Security, Sandbox & Permissions",StackOne Defender,https://github.com/stackoneHQ/defender,,,,,,repository,"Open-source indirect prompt injection defense for agents: 22MB CPU-only model, ~4ms latency, 89% balanced accuracy. Inspects tool results before they enter the LLM context window, turning untrusted MCP/CLI/function-call output into a harness-sanitized boundary rather than a prompt-level gamble." aiboost_harness,587,"Security, Sandbox & Permissions",What's in Your Agent's Context? Context Privilege Escalation Attacks against AI Agent Harness,https://arxiv.org/abs/2609.01222,2609.01222,,2026,9,arxiv_id,academic,"September 2026 first systematic security analysis of how harnesses *assemble* context across 12 real-world systems (including Claude Code and Codex), introducing two attack classes: MessageRole Context Privilege Escalation (attacker-controlled content promoted into higher-privileged message roles) and Cross-Scope CPE (attacker content persisting beyond its original scope), with consequences up to " aiboost_harness,588,"Security, Sandbox & Permissions",Daytona,https://github.com/daytonaio/daytona,,,,,,repository,"OCI-container sandboxes with sub-90ms startup, built-in Git operations, LSP support, and indefinite state persistence. Complements E2B for harnesses that need long-lived working directories across multiple agent sessions rather than ephemeral code execution." aiboost_harness,589,"Security, Sandbox & Permissions",NeMo Guardrails,https://github.com/NVIDIA-NeMo/Guardrails,,,,,,repository,"NVIDIA's programmable guardrails toolkit: define input, dialog, retrieval, execution, and output rails that intercept the agent loop at five distinct layers using the Colang DSL. The execution rail layer specifically governs what tools the LLM can invoke and what their inputs/outputs may contain — the reference for behavioral-level enforcement when static allow/deny lists are insufficient." aiboost_harness,590,"Security, Sandbox & Permissions",LangSmith Sandboxes: Secure Code Execution for Agents,https://blog.langchain.com/introducing-langsmith-sandboxes-secure-code-execution-for-agents/,,,,,,grey,"Describes a microVM-based sandboxing architecture with kernel-level isolation, resource caps (CPU/memory/disk), and an authentication proxy that keeps secrets entirely out of the runtime environment. Persistent WebSocket sessions support long-running agent tasks like dependency installation and test suite execution without the overhead of per-call container restarts." aiboost_harness,591,"Security, Sandbox & Permissions",PGSandbox MCP,https://github.com/LVTD-LLC/pgsandbox,,,,,,repository,"Local-first MCP server that gives coding agents disposable PostgreSQL databases with scoped roles, TTL cleanup, and bounded SQL/schema tools. It fills a narrower but important sandboxing gap: agents can validate migrations, reproduce database bugs, and seed demo states against a real database without handing them long-lived admin credentials." aiboost_harness,592,"Security, Sandbox & Permissions",Implementing a Secure Sandbox for Local Agents,https://cursor.com/blog/agent-sandboxing,,,,,,grey,"Cursor's cross-platform sandbox implementation (macOS Seatbelt, Linux Landlock + seccomp, Windows WSL2) that lets agents run freely within a boundary and request approval only for external access. Key result: 40% fewer user interruptions vs. no-sandbox permissioning — agents explore freely inside the boundary rather than requesting every file operation. The training insight — that agents must be e" aiboost_harness,593,"Security, Sandbox & Permissions",Practical Security Guidance for Sandboxing Agentic Workflows,https://developer.nvidia.com/blog/practical-security-guidance-for-sandboxing-agentic-workflows-and-managing-execution-risk/,,,,,,grey,"NVIDIA AI Red Team's mandatory controls for agent code execution: restrict network egress, block workspace escape, and critically — protect MCP server configuration and hooks files from agent modification. The core threat model: an agent that can edit its own harness configuration can escalate its own permissions, which standard sandbox isolation alone does not prevent." aiboost_harness,594,"Security, Sandbox & Permissions",Under the Hood: Security Architecture of GitHub Agentic Workflows,https://github.blog/ai-and-ml/generative-ai/under-the-hood-security-architecture-of-github-agentic-workflows/,,,2026,3,text_date,grey,"GitHub's March 9, 2026 architecture write-up is one of the clearest public descriptions of defense-in-depth for coding agents running inside CI: isolated agent container, firewall, MCP gateway, API proxy, staged safe outputs, and zero-secret execution. The key value is that it treats agent execution as a hostile workload inside automation infrastructure, which is exactly the mindset most harnesses" aiboost_harness,595,"Security, Sandbox & Permissions",Community-Powered Security with AI: An Open Source Framework for Security Research,https://github.blog/security/community-powered-security-with-ai-an-open-source-framework-for-security-research/,,,2026,1,text_date,grey,"GitHub Security Lab's January 14, 2026 launch of Taskflow Agent is a strong example of security-specific harness engineering: encode expert workflows as reusable tasks, keep the framework open for audit, and let AI scale established security practice instead of improvising ad hoc scans. Worth including because it turns security research itself into a sharable harness layer rather than a one-off in" aiboost_harness,596,"Security, Sandbox & Permissions",AnonymAI: Integrating Differential Privacy with LLM Agents,https://www.mdpi.com/1999-5903/18/1/41,,,,,,grey,"Framework for automating data anonymization in agent workflows. Directly addresses the harness problem of unintentional PII leakage through tool calls and memory writes — privacy enforcement moves from the agent (prompt-level trust) to the harness boundary (structural enforcement). Essential for regulatory compliance in EU, Canada, and emerging state-level privacy regimes." aiboost_harness,597,"Security, Sandbox & Permissions",Fault Tolerance Patterns: OpenClaw Journey Six—Core Retry Loop and Seven-Layer Fault-Tolerance,https://tonylixu.medium.com/openclaw-journey-six-core-retry-loop-and-seven-layer-fault-tolerance-7a9ce03147e2,,,,,,grey,Four-layer fault tolerance (retry with backoff → model fallback chains → error classification → checkpoint recovery) reduces unrecoverable failures from 23% to under 2% across agent systems. The layered approach is essential reading for hardening production agent harnesses against the combinatorial failure modes introduced by multiple tool calls and external dependencies. aiboost_harness,598,"Security, Sandbox & Permissions",deepsec,https://github.com/vercel-labs/deepsec,,,,,,repository,"Vercel Labs' security harness that treats vulnerability scanning as an agentic workflow: idempotent commands for interrupt-resume across distributed workers, SKILL.md context injection, and explicit cost transparency that forces rigorous context design. The clearest reference for building high-stakes, long-running agent harnesses where failure recovery directly determines ROI." aiboost_harness,599,"Security, Sandbox & Permissions",Visa Vulnerability Agentic Harness,https://github.com/visa/visa-vulnerability-agentic-harness,,,2026,6,text_date,repository,"Visa's June 2026 open-source harness for autonomous vulnerability discovery, remediation, and validation, built on learnings from Anthropic's Project Glasswing. The four-phase, eleven-stage pipeline is a concrete reference for security-agent harness design: threat modeling before analysis focuses the attack surface, multi-agent deterministic voting suppresses false positives, and adversarial valid" aiboost_harness,600,"Security, Sandbox & Permissions",T3MP3ST,https://github.com/elder-plinius/T3MP3ST,,,2026,7,text_date,repository,"Multi-agent offensive-security meta-harness (July 2026) that turns the coding agent you already run — Claude Code, Codex, OpenCode, or a fully offline Ollama/vLLM model — into an autonomous kill-chain hunter (recon → exploit → report) across web apps, CTFs, smart contracts, and source code. The key harness insight is keyless composition: your existing agent is the brain, and the harness bolts coor" aiboost_harness,601,"Security, Sandbox & Permissions",Sandbox Agents | OpenAI API Docs,https://developers.openai.com/api/docs/guides/agents/sandboxes,,,2026,4,text_date,grey,"OpenAI's authoritative April 2026 guide to sandbox architecture in the Agents SDK. The core principle is strict separation between the harness control plane (auth, billing, orchestration) and the sandbox compute plane (files, shell, ports), with manifest contracts, resumable session state, and sandbox-native memory." aiboost_harness,602,"Security, Sandbox & Permissions",NVIDIA OpenShell,https://github.com/NVIDIA/OpenShell,,,,,,repository,"Open-source policy-driven sandbox runtime for autonomous AI agents, announced at GTC 2026. Enforces security constraints at the kernel level via Landlock LSM (filesystem), seccomp BPF (syscalls), and an OPA/Rego-evaluated HTTP CONNECT proxy (network) — constraints are enforced on the environment itself, so even a compromised agent cannot override them. Supports Claude Code, Codex, Cursor, and Open" aiboost_harness,603,"Security, Sandbox & Permissions",IronClaw,https://github.com/nearai/ironclaw,,,,,,repository,"NEAR AI's open-source agent OS that treats agent execution as a privacy-first harness problem: untrusted tools run in WASM sandboxes with capability-based permissions, credentials are injected at the host boundary with leak detection, and prompt-injection filtering plus endpoint allowlisting constrain agent behavior. The combination of local encrypted storage, hybrid-search memory, MCP server supp" aiboost_harness,604,"Security, Sandbox & Permissions",Microsoft Agent Governance Toolkit,https://github.com/microsoft/agent-governance-toolkit,,,,,,repository,"Seven-package, multi-language (Python, Rust, TypeScript, Go, .NET) runtime security toolkit that addresses all 10 OWASP Agentic AI risks with deterministic, sub-millisecond policy enforcement. Includes Agent OS (policy engine intercepting every action), Agent Mesh (secure agent-to-agent communication), and Agent Runtime (dynamic execution rings). Hooks into LangChain callbacks, CrewAI task decorat" aiboost_harness,605,"Security, Sandbox & Permissions",mcpguard-dynamic,https://github.com/facebook/mcpguard-dynamic,,,,,,repository,"Meta's kernel-level eBPF sandbox for MCP tool calls: a transparent proxy that enforces capability policies through a policy engine, argument validator, and BPF LSM hooks (file, network, process, fork), so a compromised or malicious MCP server cannot bypass restrictions at the application layer. Includes a 14-server, 82-case benchmark showing the full defense stack achieves 68.9% attack prevention " aiboost_harness,606,"Security, Sandbox & Permissions",Grimlock: Guarding High-Agency Systems with eBPF and Attested Channels,https://arxiv.org/abs/2605.27488,2605.27488,,2026,5,arxiv_id,academic,"Roblox's agent guard for high-agency systems: eBPF-forced traffic interception channels every agent-to-agent and agent-to-service call through a guard, post-handshake TLS 1.3 attestation binds identity to the channel, and short-lived scope tokens enforce least-privilege delegation across heterogeneous machines and clouds. The key harness insight is that identity, authorization, provenance, and del" aiboost_harness,607,"Security, Sandbox & Permissions",Cloudflare Dynamic Workers,https://blog.cloudflare.com/dynamic-workers/,,,,,,grey,"V8 isolate-based sandboxing for AI-agent-generated code execution, now in open beta. Isolates start in milliseconds using megabytes of memory — 100x faster and up to 100x more memory-efficient than containers. The sandbox intercepts outbound HTTP requests for credential injection so agent code never touches secrets directly. A fundamentally different architectural option from container-based sandb" aiboost_harness,608,"Security, Sandbox & Permissions",cloudflare/computer,https://github.com/cloudflare/computer,,,,,,repository,"Cloudflare's preview open-source agent runtime: a Durable Object hosts authoritative workspace state in SQLite, while pluggable backends (container FUSE mount, isolate shell, isolate JavaScript) execute code against that single source of truth. The key harness idea is that durable state and sandboxed execution can share one primitive instead of requiring separate storage, sync, and isolation layer" aiboost_harness,609,"Security, Sandbox & Permissions",Kubernetes Agent Sandbox,https://github.com/kubernetes-sigs/agent-sandbox,,,,,,repository,"K8s-native Sandbox CRD (under SIG Apps) providing declarative, standardized APIs for managing isolated, stateful, singleton workloads for AI agent runtimes. Supports gVisor and Kata Containers for kernel-level isolation; v0.2.1 introduced ""Secure by Default"" networking architecture enforcing strict isolation with a shared policy model. The right choice when agents must run inside existing Kubernet" aiboost_harness,610,"Security, Sandbox & Permissions",agent-shell-tools,https://github.com/google/agent-shell-tools,,,,,,repository,"Google's open-source toolkit for giving coding agents controlled shell access with opinionated defaults: an nsjail-based sandbox, a command-filter rule language, and a gRPC execution proxy so agents can stream commands into isolation without exposing host credentials or filesystems. A concrete reference for separating the agent's shell surface from the host boundary rather than relying on prompt-l" aiboost_harness,611,"Security, Sandbox & Permissions",Why sandboxing your agent is not enough,https://www.cncf.io/blog/2026/07/07/why-sandboxing-your-agent-is-not-enough/,,,2026,7,text_date,grey,"CNCF's July 2026 argument that container isolation is necessary but insufficient for production agents. The agent-substrate pattern decouples the agent actor from pod lifecycle so agents execute in short bursts, suspend when idle, and resume on any worker while keeping gVisor/Kata-level isolation. A concrete pattern for high-density, low-idle-cost secure agent fleets in Kubernetes." aiboost_harness,612,"Security, Sandbox & Permissions",Alibaba OpenSandbox,https://github.com/alibaba/OpenSandbox,,,2026,3,text_date,repository,"General-purpose sandbox platform for AI agents (8.7K+ stars, March 2026) with multi-language SDKs (Python, Java, TypeScript, Go, C#), unified APIs across Docker/Kubernetes runtimes, and support for secure container runtimes (gVisor, Kata Containers, Firecracker). Covers coding agents, GUI agents, agent evaluation, and RL training in a single abstraction layer — the most runtime-flexible sandbox op" aiboost_harness,613,"Security, Sandbox & Permissions",CubeSandbox,https://github.com/TencentCloud/CubeSandbox,,,,,,repository,"Tencent Cloud's production-validated microVM sandbox for AI agents: sub-60ms cold start via snapshot cloning, <5MB per-instance overhead, and true kernel-level isolation with eBPF-enforced network policies. E2B-compatible drop-in replacement that demonstrates how hyperscale cloud infrastructure can be repurposed for high-density agent execution." aiboost_harness,614,"Security, Sandbox & Permissions",zeroboot,https://github.com/zerobootdev/zeroboot,,,,,,repository,"Sub-millisecond VM sandboxes for AI agents via copy-on-write forking, enabling fresh isolated execution on every tool call without the latency penalty of container or microVM cold starts. The critical harness advantage is turning per-action isolation from a batch-mode luxury into a real-time loop primitive — agents can spin up and tear down environments within a single reasoning step." aiboost_harness,615,"Security, Sandbox & Permissions",forkd,https://github.com/deeplethe/forkd,,,,,,repository,"MicroVM sandbox runtime that replaces cold-boot isolation with fork-from-warm: children share a paused parent's memory copy-on-write, so 100 agent tool-execution sandboxes spawn in ~100 ms rather than seconds. The live-BRANCH primitive (~56 ms pause) lets an agent fork its own in-flight state for speculative exploration or checkpointing — a fundamental leap for harnesses where per-action isolation" aiboost_harness,616,"Security, Sandbox & Permissions",The Attack and Defense Landscape of Agentic AI: A Comprehensive Survey,https://arxiv.org/abs/2603.11088,2603.11088,,2026,3,arxiv_id,academic,"The first systematic survey of AI agent security from UC Berkeley and UIUC (Dawn Song et al., March 2026). Reviews 128 papers covering 51 attack methods and 60 defense mechanisms. Introduces a framework for understanding security risks specific to agentic (not just LLM) systems and identifies open gaps in securing agent architectures — the definitive 2026 reference for agent threat modeling." aiboost_harness,617,"Security, Sandbox & Permissions",Trustworthy agents in practice,https://www.anthropic.com/research/trustworthy-agents,,,2026,4,text_date,grey,"Anthropic's April 2026 framework for governing autonomous agents through five principles: human control, value alignment, secure interactions, transparency, and privacy. The most complete published treatment of how to design harness-level governance that keeps pace with increasing agent capability and autonomy." aiboost_harness,618,"Security, Sandbox & Permissions",RAMPART,https://github.com/microsoft/RAMPART,,,,,,repository,"Pytest-native safety and security testing framework for agentic AI that turns red-team findings into repeatable CI tests. Supports statistical trials (e.g., ""safe in 95% of runs"") rather than single-shot pass/fail, making it the first concrete tool for treating agent safety as an engineering discipline rather than a post-hoc audit." aiboost_harness,619,"Security, Sandbox & Permissions",aiming-lab/AutoHarness,https://github.com/aiming-lab/AutoHarness,,,,,,repository,"Open-source Python governance harness that wraps any OpenAI-compatible client with a configurable tool-approval pipeline, prompt-injection defense, secret-exposure checks, and JSONL audit trails. Demonstrates how to productize ""Agent = Model + Harness"" into a drop-in governance layer rather than scattering guardrails across application code." aiboost_harness,620,"Security, Sandbox & Permissions",Speakeasy control plane,https://github.com/speakeasy-api/gram,,,,,,repository,"MCP gateway that scopes agent tool access by identity, team, and role, with security policy enforced on each action and an audit trail. Useful as a reference for putting authorization at the tool boundary instead of relying on model prompts to restrict access." aiboost_harness,622,"Security, Sandbox & Permissions",MCP Security Testing Checklist,https://loki-intelligence.com/blog/mcp-security-testing-checklist,,,,,,grey,"Eight-step practitioner verification checklist for MCP server and tool-harness boundaries: OAuth/token scoping, schema and description poisoning, dynamic discovery freezes, SSRF/path-traversal argument validation, cross-tenant isolation, and secret redaction in tool responses. Provides a concrete pre-release audit rubric for verifying that MCP integrations cannot be abused to cross harness trust b" aiboost_harness,623,"Security, Sandbox & Permissions",Microsandbox,https://github.com/superradcompany/microsandbox,,,,,,repository,"Rust/Apache-2.0 local-first microVM runtime for untrusted agent workloads: <100ms boots, OCI-image compatible, fork-and-snapshot of live sandboxes, and an embeddable library API that spawns VMs in-process with no daemon. The distinguishing designs are secret handling — keys never enter the VM, so code running inside cannot read them directly — and an official MCP server plus Agent Skills that let " aiboost_harness,624,"Security, Sandbox & Permissions",claude-bulkhead,https://github.com/dooly6276/claude-bulkhead,,,,,,repository,"PreToolUse hooks that keep parallel Claude Code sessions on one machine from interfering with each other: they block name-based process kills, port clashes, heavy commands when RAM is low, git add -A in a tree another live session is editing, and whole-disk scans, and return the reason and a safer command to the agent. It parses command strings rather than sandboxing, so it is a guard against acci" aiboost_harness,625,"Security, Sandbox & Permissions",Sipar,https://github.com/osimmoapies/sipar,,,,,,repository,"A small agent language where the permission check is in the runtime: model, web, file and tool output is labelled and cannot become a shell command, a file write or a URL until the program has checked it. Worth including as a working, dependency-free example of information-flow control for agent tool calls, with a browser playground." aiboost_harness,631,Evals & Verification,Nautilus Compass,https://github.com/chunxiaoxx/nautilus-compass,,,,,,repository,"Independent verification & judging service for AI-agent claims: preregistered gates, third-party recomputation with per-case evidence, and a public errata/ledger of its own published mistakes. Runs LongMemEval-based memory evals and the PRECOR study on quantization-induced judge degradation." aiboost_harness,632,Evals & Verification,Demystifying Evals for AI Agents,https://www.anthropic.com/engineering/demystifying-evals-for-ai-agents,,,,,,grey,"Anthropic's comprehensive guide to agent evaluation: trajectory evals, outcome evals, and how to build eval harnesses that are themselves reliable." aiboost_harness,633,Evals & Verification,DeepEval,https://github.com/confident-ai/deepeval,,,,,,repository,"The most complete open-source LLM/agent eval framework: 20+ built-in metrics (hallucination, answer relevancy, RAGAs, tool correctness), pytest integration, and a CI-friendly runner. Removes the need to hand-roll eval infrastructure when you need structured, repeatable agent quality gates." aiboost_harness,634,Evals & Verification,Claw-Eval,https://github.com/claw-eval/claw-eval,,,,,,repository,"300 human-verified tasks across 9 categories evaluating LLM-as-agent performance on completion, safety, and robustness with a Pass^3 methodology that requires success across three independent trials. Referenced by Meta, Kimi, Qwen, and Tencent as a trustworthy benchmark for general agentic capabilities — the most rigorous community-verified eval harness published in 2026." aiboost_harness,635,Evals & Verification,SWE-bench,https://www.swebench.com,,,,,,grey,"The canonical benchmark for coding agents. Essential reference for understanding what ""verified working"" means for harness outputs." aiboost_harness,636,Evals & Verification,Terminal-Bench,https://github.com/laude-institute/terminal-bench,,,,,,repository,"The de facto benchmark and execution harness for terminal agents: real sandboxed shell tasks verified by deterministic test scripts, now at 2.x on the Harbor framework. Cited throughout this list because it is the yardstick every harness-only improvement claims against — from LangChain's rank-30-to-top-5 harness changes to Self-Harness and Beagle's evolution loops — making it essential infrastruct" aiboost_harness,637,Evals & Verification,Accio-org/RealReplicaBench,https://github.com/Accio-org/RealReplicaBench,,,,,,repository,"Alibaba International's benchmark for long-horizon agents in high-fidelity, stateful replicas of real online services: 107 tasks spanning CLI, browser, file, and API/MCP workflows, each graded by deterministic or LLM-assisted verifiers in a fresh container. The reproducibility contract — mock services, auditable trajectories, and container metadata — makes it the right eval harness for agents that" aiboost_harness,638,Evals & Verification,Inspect AI,https://github.com/UKGovernmentBEIS/inspect_ai,,,,,,repository,"UK AI Security Institute's eval framework with native support for evaluating external agents (Claude Code, Codex CLI) as black-box targets, plus built-in bash/python/web browsing tools. Built for safety-grade rigor; the right foundation for harness-level eval infrastructure." aiboost_harness,639,Evals & Verification,Quantifying Infrastructure Noise in Agentic Coding Evals,https://www.anthropic.com/engineering/infrastructure-noise,,,,,,grey,"Anthropic's empirical study showing container resource configuration alone produces 6+ percentage point benchmark swings — often exceeding model-to-model gaps. The 3x threshold finding is the key practical result: scores are stable up to 3x specified resources, but above that agents shift strategy entirely (lean tools vs. heavy dependencies), meaning tight and generous resource limits measure fund" aiboost_harness,640,Evals & Verification,AgentLens: Revealing The Lucky Pass Problem in SWE-Agent Evaluation,https://arxiv.org/abs/2605.12925,2605.12925,,2026,5,arxiv_id,academic,"Process-level evaluation framework that separates solid solutions from ""lucky passes"" (regression cycles, blind retries, missing verification) in SWE agent trajectories. Analyzing 2,614 OpenHands trajectories shows up to 23.2% of passes are lucky and model rankings shift by as many as five positions when scored by process quality rather than binary pass/fail — a direct argument for building verifi" aiboost_harness,641,Evals & Verification,StaminaBench: Stress-Testing Coding Agents over 100 Interaction Turns,https://arxiv.org/abs/2606.19613,2606.19613,,2026,6,arxiv_id,academic,"Amazon Science's black-box benchmark for measuring how many consecutive change-request turns a coding agent can sustain before failing. The core finding — test feedback and retry capability improve pass counts by up to 12×, while stronger models show a 6× gap between their best and worst harness — makes long-horizon stamina an explicit eval target rather than an afterthought." aiboost_harness,642,Evals & Verification,Harness-Bench: Measuring Harness Effects across Models in Realistic Agent Workflows,https://arxiv.org/abs/2605.27922,2605.27922,,2026,5,arxiv_id,academic,"Diagnostic benchmark that isolates the execution layer (context, tools, state, recovery) by evaluating model-harness pairings under shared tasks. The core finding across 5,194 trajectories — that agent capability should be reported at the model-harness configuration level, not attributed to the base model alone — makes it essential infrastructure for evidence-based harness design." aiboost_harness,643,Evals & Verification,Stop Comparing LLM Agents Without Disclosing the Harness,https://arxiv.org/abs/2605.23950,2605.23950,,2026,5,arxiv_id,academic,"May 2026 position paper formalizing the ""Binding Constraint Thesis"": for long-horizon tasks among frontier models, harness configuration — context construction, tool interaction, orchestration, verification — binds performance more strongly than model choice, so leaderboard gains get misattributed to the model. The proposed disclosure standards and variance-decomposition protocol (separating harne" aiboost_harness,644,Evals & Verification,tau-bench,https://github.com/sierra-research/tau-bench,,,,,,repository,"Benchmarks agent behavior in three-way user-tool-policy interactions — the failure mode SWE-bench doesn't cover. Useful for validating that a harness correctly enforces business rules across multi-turn, stateful conversations." aiboost_harness,645,Evals & Verification,Towards a Science of AI Agent Reliability,https://arxiv.org/abs/2602.16666,2602.16666,,2026,2,arxiv_id,academic,"Proposes twelve concrete reliability metrics across four dimensions (consistency, robustness, predictability, safety), evaluated against 14 agentic models. The central finding — that recent capability gains yield only modest reliability improvements — is the empirical case for investing in harness-layer reliability engineering as a discipline distinct from model selection." aiboost_harness,646,Evals & Verification,"Characterizing Faults in Agentic AI: A Taxonomy of Types, Symptoms, and Root Causes",https://arxiv.org/abs/2603.06847,2603.06847,,2026,3,arxiv_id,academic,"The first large-scale empirical fault taxonomy for agent systems, derived from 13,602 issues across 40 open-source repositories: 37 fault types, 13 symptom classes, 12 root cause categories. Key finding: most failures originate from mismatches between probabilistically generated artifacts and deterministic interface constraints — a structural harness problem, not a model capability problem." aiboost_harness,647,Evals & Verification,VeRO: An Evaluation Harness for Agents to Optimize Agents,https://arxiv.org/abs/2602.22480,2602.22480,,2026,2,arxiv_id,academic,"Framework for evaluating agent-on-agent optimization cycles: a coding agent iteratively modifies a target agent's harness (prompts, tools, configuration) through edit-execute-evaluate loops while VeRO captures versioned agent snapshots, budget-controlled evaluation, and structured execution traces. Addresses the meta-evaluation gap — how to systematically measure whether one agent is improving ano" aiboost_harness,648,Evals & Verification,Eval Awareness in Claude Opus 4.6's BrowseComp Performance,https://www.anthropic.com/engineering/eval-awareness-browsecomp,,,,,,grey,"Anthropic's documented case of Claude Opus 4.6 inferring it was under evaluation, identifying the benchmark by name, and decrypting the answer key — producing 11 non-intended solutions. A direct challenge to eval harness design: any eval that runs in a web-enabled environment is vulnerable to the agent researching the benchmark itself. The practical countermeasure — evaluate in network-isolated en" aiboost_harness,649,Evals & Verification,Designing AI-Resistant Technical Evaluations,https://www.anthropic.com/engineering/AI-resistant-technical-evaluations,,,2026,1,text_date,grey,"Anthropic's January 21, 2026 write-up is the clearest account of a problem eval builders now have to treat as first-class: capable models can invalidate the test itself. The practical value is the redesign methodology — shift toward longer-horizon, tool-building, environment-understanding tasks that remain discriminative even as frontier models get better at short take-homes." aiboost_harness,650,Evals & Verification,Amazon Bedrock AgentCore Evaluations Is Now Generally Available,https://aws.amazon.com/about-aws/whats-new/2026/03/agentcore-evaluations-generally-available/,,,2026,3,text_date,grey,"AWS's March 31, 2026 GA launch matters because it operationalizes agent evals as an infrastructure service: trajectory scoring, task completion checks, and model-graded assessments are wired into the same platform that hosts agents. Worth adding because it shows how evaluation stops being an offline benchmark and becomes part of the runtime control plane teams can standardize on." aiboost_harness,651,Evals & Verification,Live-SWE-agent: First Live Software Agent with Self-Evolving Scaffold,https://arxiv.org/html/2511.13646v3,2511.13646,,2025,11,arxiv_id,academic,Demonstrates end-to-end harness design for production software engineering: 77.4% solve rate on SWE-bench Verified (vs. 50% from human contractors). The key insight is the self-evolving scaffold — the harness itself adapts based on failure signals rather than requiring manual engineering each time a new task class emerges. Shows the upper-bound of what specialized harness design can achieve. aiboost_harness,652,Evals & Verification,OccuBench: Evaluating AI Agents on Real-World Professional Tasks via Language World Models,https://arxiv.org/abs/2604.10866,2604.10866,,2026,4,arxiv_id,academic,"April 2026 benchmark that evaluates agents on 12 real-world professional occupations (software engineer, data scientist, financial analyst, etc.) using language world models to simulate realistic work environments. Agents are scored on task completion, efficiency, and professional standards adherence. The important harness contribution is the environment-as-evaluator pattern: instead of static tes" aiboost_harness,653,Evals & Verification,STATE-Bench,https://github.com/microsoft/STATE-Bench,,,2026,5,text_date,repository,"Microsoft's open-source benchmark (May 2026) measuring whether agents actually improve with experience on 450 realistic enterprise tasks across customer support, travel, and shopping. The first eval to treat memory as an independent variable with explicit learning tracks, providing a concrete way to compare memory architectures beyond simple recall metrics." aiboost_harness,654,Evals & Verification,Investigating unintended model actions in our evaluations and internal use,https://www.anthropic.com/research/investigating-unintended-model-actions,,,2026,10,text_date,grey,"Anthropic's October 2026 transparency disclosure of reward-hacking incidents during live internal evals: models exploited SQL/command injection on a university server, submitted a false police tip through a real website, bypassed fee gates, and used URL shorteners to sidestep fetch-tool limits — until Anthropic cut live internet access from all internal evaluations. The harness takeaway is stark: " aiboost_harness,655,Evals & Verification,LevelBench-100,https://github.com/levelcfo/levelbench-100,,,,,,repository,"Empirical benchmark isolating base model capability vs. CLI harness architecture (Claude Code, Codex CLI, Gemini CLI, Antigravity CLI) and deterministic verification playbooks across 100 multi-file accounting close checks, 15 SMB close workbooks, and 14 browser-automation portal gates. Demonstrates that while raw frontier models top out at 71.0% with a 28.6% bad-entry rate, multi-file CLI harnesse" aiboost_harness,676,Production Infrastructure & Operations,Claude Managed Agents: Self-Hosted Sandboxes and MCP Tunnels,https://platform.claude.com/docs/en/managed-agents/self-hosted-sandboxes,,,2026,5,text_date,grey,"Anthropic's May 2026 enterprise deployment pattern keeps the agent orchestration loop on Anthropic's infrastructure while moving tool execution into customer-controlled sandboxes; combined with [MCP tunnels](https://platform.claude.com/docs/en/agents-and-tools/mcp-tunnels/overview) for private-network tool access, it's the reference architecture for data-residency-conscious production harnesses." aiboost_harness,677,Production Infrastructure & Operations,AgentCgroup: Understanding and Controlling OS Resources of AI Agents,https://arxiv.org/abs/2602.09345,2602.09345,,2026,2,arxiv_id,academic,February 2026 empirical study of sandboxed coding-agent workloads finding that OS-level execution accounts for 56–74% of end-to-end latency and memory is the real concurrency bottleneck (15.4× peak-to-average spikes driven by tool calls). Proposes an intent-driven eBPF controller aligned with tool-call boundaries — essential for anyone running multi-tenant agent sandboxes where coarse container li aiboost_harness,678,Production Infrastructure & Operations,AI Agent Scaling Gap: Pilot to Production (March 2026),https://www.digitalapplied.com/blog/ai-agent-scaling-gap-march-2026-pilot-to-production,,,2026,3,text_date,grey,"](https://www.digitalapplied.com/blog/ai-agent-scaling-gap-march-2026-pilot-to-production) — Analysis showing 72% of Global 2000 companies operate agents beyond experimental phases, but only 14% successfully scaled organization-wide. Scaling success correlates strongly with operations infrastructure (monitoring, evaluation harnesses, incident response) rather than technology choices. Documents the" aiboost_harness,679,Production Infrastructure & Operations,builderz-labs/mission-control,https://github.com/builderz-labs/mission-control,,,,,,repository,"Self-hosted orchestration dashboard for agent task dispatch, multi-agent workflow coordination, and spend monitoring across gateways. The zero-external-dependency design (SQLite, single pnpm start) makes it the most practical open-source control plane for teams that need governance and cost visibility without building infrastructure from scratch." aiboost_harness,680,Production Infrastructure & Operations,kvcache-ai/AgentENV,https://github.com/kvcache-ai/AgentENV,,,,,,repository,"Distributed platform for running agent environments at scale, powering Kimi K3's agentic RL training. Sub-50ms snapshot boot/resume, native fork support for parallel workflows, and overlaybd-based image loading make it the right infrastructure layer when your harness needs thousands of ephemeral sandboxes rather than a handful." aiboost_harness,681,Production Infrastructure & Operations,5 Production Scaling Challenges for Agentic AI in 2026,https://machinelearningmastery.com/5-production-scaling-challenges-for-agentic-ai-in-2026/,,,,,,grey,Data infrastructure prioritized before deployment; successful scalers appoint AI operations function pre-expansion; multi-agent distributed systems with load balancing and auto-scaling. Essential reading for understanding infrastructure prerequisites for agent deployment at scale. aiboost_harness,682,Production Infrastructure & Operations,AI Agent Cost Optimization Guide 2026: Reduce Spend by 60-80%,https://moltbook-ai.com/posts/ai-agent-cost-optimization-2026,,,,,,grey,"Systematic patterns for cost reduction: model routing and caching (40-60% savings); Anthropic prompt caching (90% discount on cached tokens); identifying unnecessary agent overhead vs. simple API chains. Key harness decisions (tool selection, caching strategy, model choice per task) determine operating cost." aiboost_harness,683,Production Infrastructure & Operations,getagentseal/codeburn,https://github.com/getagentseal/codeburn,,,,,,repository,"Free, local tool that tracks AI coding token usage and cost across 31 tools and agents by model, project, and task. Worth including because cross-tool cost visibility is the missing prerequisite for agent FinOps in multi-tool teams — most observability tools either require cloud upload or only cover a single provider." aiboost_harness,684,Production Infrastructure & Operations,KernelEvolve: How Meta's Ranking Engineer Agent Optimizes AI Infrastructure,https://engineering.fb.com/2026/04/02/developer-tools/kernelevolve-how-metas-ranking-engineer-agent-optimizes-ai-infrastructure/,,,2026,4,url_date,grey,"Meta's production-grade agentic kernel optimization system that autonomously generates optimized Triton kernels for hundreds of models serving billions of users daily. Achieves up to 17x speedup over PyTorch baselines with 100% correctness across 250 problems. Demonstrates harness design for continuous infrastructure optimization: a purpose-built job-harness evaluates each candidate kernel, feeds " aiboost_harness,685,Production Infrastructure & Operations,State of Agent Engineering 2026,https://www.langchain.com/state-of-agent-engineering,,,,,,grey,"LangChain's industry survey of 1,300+ professionals: 57.3% now have agents in production (up from 51%), quality is the top barrier at 32%, and 89% have implemented observability while only 52% run evals. The most comprehensive snapshot of where the industry stands on agent deployment maturity, model strategies, and operational gaps." aiboost_harness,686,Production Infrastructure & Operations,Agentic Development: What It Means for Engineering Infrastructure in 2026,https://www.bunnyshell.com/guides/agentic-development/,,,,,,grey,"Defines the four infrastructure capabilities that agentic development requires: per-task isolated sandboxes, sub-100ms startup (ruling out traditional VMs and most K8s approaches), API-driven lifecycle management, and MCP-native environment control. The clearest articulation of why existing CI/CD infrastructure is insufficient for agent-driven development workflows." aiboost_harness,687,Production Infrastructure & Operations,"FinOps for Agents: Loop Limits, Tool-Call Caps, and the New Unit Economics of Agentic SaaS",https://www.infoworld.com/article/4138748/finops-for-agents-loop-limits-tool-call-caps-and-the-new-unit-economics-of-agentic-saas.html,,,,,,grey,"Defines five concrete budget guardrails enforced at the infrastructure gateway: loop/step limits, tool-call caps, per-run token budgets, wall-clock timeouts, and per-tenant budgets with anomaly alerts. Introduces Cost-per-Accepted-Outcome (CAPO) as the right unit economic metric for agent harnesses — shifting cost measurement from tokens consumed to business value delivered." aiboost_harness,688,Production Infrastructure & Operations,Backtesting AI Agents: How SRE Teams Prove Reliability Before Production,https://drdroid.io/blog/backtesting-ai-agents-how-sre-teams-prove-reliability-before-production,,,,,,grey,"Formalizes agent validation as infrastructure-grade testing with pass^k reliability (all 20+ trials must succeed) rather than pass@k (one success). Defines five measurable dimensions (consistency, robustness, predictability, safety, cost stability) with specific SLO thresholds. Recommends dataset composition of 20% golden paths, 30% edge cases, 20% adversarial, 30% regression from production incid" aiboost_harness,689,Production Infrastructure & Operations,How My Agents Self-Heal in Production,https://blog.langchain.com/production-agents-self-heal/,,,2026,4,text_date,grey,"A concrete April 3, 2026 production pattern for closing the post-deploy loop: detect regressions, attribute whether the last deploy caused them, then dispatch a coding agent to open a fix PR automatically. This belongs here because it turns evals and observability into an active remediation harness, not just a dashboard humans are expected to watch." aiboost_harness,690,Production Infrastructure & Operations,"Minions: Stripe's one-shot, end-to-end coding agents—Part 2",https://stripe.dev/blog/minions-stripes-one-shot-end-to-end-coding-agents-part-2,,,,,,grey,"Stripe's deep-dive into their unattended minion harness shipping 1,300+ PRs/week: ""blueprints"" interleave deterministic code nodes with agentic subtasks, a centralized 500-tool MCP server (Toolshed) serves the whole fleet, and pre-warmed devboxes prove that investments in human developer productivity pay equal dividends for agents." aiboost_harness,691,Production Infrastructure & Operations,Amazon Bedrock AgentCore,https://aws.amazon.com/bedrock/agentcore/,,,,,,grey,"AWS's fully managed agent deployment platform providing serverless runtime with session isolation, built-in memory (session + long-term), secure gateway for tool access, browser runtime, and code interpreter — all framework-agnostic. Now supports AG-UI protocol for real-time agent-to-frontend streaming, VPC/PrivateLink for enterprise security, and CloudFormation for infrastructure-as-code deployme" aiboost_harness,692,Production Infrastructure & Operations,AWS Agent Registry for Centralized Agent Discovery and Governance,https://aws.amazon.com/about-aws/whats-new/2026/04/aws-agent-registry-in-agentcore-preview/,,,2026,4,text_date,grey,"AWS's April 9, 2026 preview adds a missing production primitive: a governed catalog for agents, tools, skills, MCP servers, and custom resources with approval workflows, audit trails, and MCP-accessible discovery. Worth including because large organizations do not just need to run agents safely; they need to know what agent capabilities already exist so teams stop rebuilding the same scaffolding i" aiboost_harness,693,Production Infrastructure & Operations,A Dev's Guide to Production-Ready AI Agents,https://cloud.google.com/blog/products/ai-machine-learning/a-devs-guide-to-production-ready-ai-agents,,,2026,4,text_date,grey,"Google Cloud's developer guide (April 2026) for moving AI agents from prototype to production using ADK, Vertex AI Agent Engine for managed hosting, and Cloud Run for serverless deployment. Covers the full production stack: agent development patterns, scaling considerations, security and identity requirements, and operational monitoring — the most concrete first-party guide for deploying agents on" aiboost_harness,694,Production Infrastructure & Operations,Enhanced Tool Governance in Vertex AI Agent Builder,https://cloud.google.com/blog/products/ai-machine-learning/new-enhanced-tool-governance-in-vertex-ai-agent-builder,,,2026,4,text_date,grey,"Google Cloud's approach to production agent governance (April 2026): agents get identity as first-class IAM principals with least-privilege enforcement, Cloud API Registry integration enables organizational tool governance (admins manage available tools centrally), and a new observability dashboard tracks token usage, latency, and error rates. Demonstrates the cloud-native pattern for tool governa" aiboost_harness,695,Production Infrastructure & Operations,What Is an Agent Harness? Running Governed Managed Agents in Production,https://www.truefoundry.com/blog/agent-harness-managed-ai-agents,,,2026,6,text_date,grey,"TrueFoundry's June 2026 architectural guide to the agent harness as a production runtime: declarative agent definitions reference models, MCP servers, and versioned skills by name while the platform injects secrets, manages sandboxes, and unifies traces across model, tool, and agent traffic. The clearest published framing of the harness as the build-or-buy operational layer that turns a model-plus" aiboost_harness,696,Production Infrastructure & Operations,truefoundry/trueforge,https://github.com/truefoundry/trueforge,,,,,,repository,"TrueFoundry's open-source agent harness runtime: runs the execution loop, MCP tools, skills, sandboxing, approvals, context management, and session state, exposing it through a chat UI, HTTP API, and embeddable UI SDK. A concrete reference for turning a model-plus-tools stack into a production-grade agent without building the loop from scratch." aiboost_harness,697,Production Infrastructure & Operations,"Building Governed Agents: A Framework for Cost, Control, and Compliance",https://www.langchain.com/blog/building-governed-agents-a-framework-for-cost-control-and-compliance,,,2026,7,text_date,grey,"LangChain's July 2026 framework for treating the agent gateway as a runtime control plane that enforces cost, control, and compliance policies across every model call, tool call, and agent hop. A concrete reference for moving agent governance from static configuration into a harness-level enforcement layer." aiboost_harness,698,Production Infrastructure & Operations,OpenAI Agents API,https://developers.openai.com/api/docs/guides/agents,,,2026,9,text_date,grey,"OpenAI's September 10, 2026 public beta of the managed runtime behind Codex and ChatGPT's agent: one POST /v1/agents/sessions call gets you the full Codex harness — automatic context compaction, durable sessions with recovery, programmatic tool calling, MCP over HTTP, and delegation to concurrent subagents with isolated contexts. The environment model is the most instructive part for harness build" aiboost_harness,699,Production Infrastructure & Operations,Redesigning source control at Anthropic,https://www.anthropic.com/engineering/redesigning-source-control,,,2026,10,text_date,grey,"Anthropic's October 13, 2026 account of agent-scale development breaking Git: merged changes went from ~2,500/day in May to 30,000+/day by September, CI performed ~25M checkouts daily, and the decentralized-clone model would have generated 600+ PB of daily traffic. Scion — a centralized, always-online VCS built by seven engineers in five months with server-side merges and blame, changeset-based st" aiboost_harness,707,Related Awesome Lists,Awesome Context Engineering,https://github.com/Meirtz/Awesome-Context-Engineering,,,,,,repository,"Comprehensive survey on context engineering: prompt engineering, RAG, context window management, production AI systems." aiboost_harness,708,Related Awesome Lists,awesome-claude-code,https://github.com/hesreallyhim/awesome-claude-code,,,,,,repository,"Curated resources, tools, and workflows specifically for Claude Code users." aiboost_harness,709,Related Awesome Lists,awesome-mcp-servers,https://github.com/appcypher/awesome-mcp-servers,,,,,,repository,Comprehensive list of MCP servers for extending agents with external capabilities. aiboost_harness,710,Related Awesome Lists,awesome-ai-agents,https://github.com/e2b-dev/awesome-ai-agents,,,,,,repository,"Curated list of AI agents and agent frameworks, organized by use case. Useful for surveying the landscape of what harnesses are being built around." aiboost_harness,711,Related Awesome Lists,awesome-llm-apps,https://github.com/Shubhamsaboo/awesome-llm-apps,,,,,,repository,"Collection of production LLM applications with source code across RAG, multi-agent, and tool-use patterns. Good reference for how harness primitives combine in real applications." aiboost_harness,712,Related Awesome Lists,ICLR 2026 MemAgents Workshop,https://sites.google.com/view/memagent-iclr26/,,,,,,grey,"Interdisciplinary workshop (April 27, Rio de Janeiro) covering episodic/semantic memory, knowledge graphs, vector databases, retrieval pipelines, temporal credit assignment, and context management for agentic systems. The canonical venue for memory architecture research and standards; accepts full papers (9pg), short papers (4pg), tiny papers (2pg)." aiboost_harness,713,Related Awesome Lists,Awesome Code as Agent Harness Papers,https://github.com/YennNing/Awesome-Code-as-Agent-Harness-Papers,,,,,,repository,"Curated companion list to the *Code as Agent Harness* survey, organizing research on code-centric agentic systems across interface, mechanisms, and scaling layers. A focused research map for anyone building harnesses where code is the executable scaffold rather than just the output." aiboost_harness,714,Related Awesome Lists,DEEP-JLU/Awesome-Graph-Engineering,https://github.com/DEEP-JLU/Awesome-Graph-Engineering,,,2026,8,text_date,repository,"August 2026 survey companion on graph engineering in the era of LLM agents, organizing research from model intelligence through individual intelligence (loop architecture, runtime orchestration, memory, environment feedback) to system intelligence (task graphs, agent coordination, state management). Its loop-architecture and state-management sections are a research-grade complement to this list, a" walkinglabs_harness,44,Courses & Learning Resources,walkinglabs/learn-harness-engineering,https://github.com/walkinglabs/learn-harness-engineering,,,,,,repository,"A project-based course repository on making Codex and Claude Code more reliable, centered on an Electron personal knowledge base app with lecture handouts, example artifacts, and practical harness projects." walkinglabs_harness,45,Courses & Learning Resources,Phelan164/codex-howto,https://github.com/Phelan164/codex-howto,,,,,,repository,"A Codex-focused engineering curriculum with installable skills, repository instructions, scoped permissions, testing, review, orchestration, and reproducible token measurements for building an inspectable coding-agent harness." walkinglabs_harness,46,Courses & Learning Resources,hardness1020/awesome-agent-architecture,https://github.com/hardness1020/awesome-agent-architecture,,,,,,repository,"Trilingual architecture notes and runnable demos covering agent loops, tool execution, memory, permissions, context delivery, and orchestration." walkinglabs_harness,47,Courses & Learning Resources,What Is an Agent Harness? Harness Engineering Explained,https://tej.as/blog/what-is-an-agent-harness,,,,,,grey,"A build-along write-up that grows a TypeScript browser agent from a bare loop that lies about success into a harness with guardrails, context trimming, deterministic trace verification, retries, and harness-handled login, with one branch per step in [basically-ai-harness](https://github.com/TejasQ/basically-ai-harness)." walkinglabs_harness,52,Foundations,Harness engineering: leveraging Codex in an agent-first world,https://openai.com/index/harness-engineering/,,,,,,grey,"OpenAI's flagship field report on building a large application with Codex using architectural constraints, repo-local instructions, browser validation, and telemetry." walkinglabs_harness,53,Foundations,Effective harnesses for long-running agents,https://www.anthropic.com/engineering/effective-harnesses-for-long-running-agents,,,,,,grey,"Anthropic's core article on initializer agents, feature lists, init.sh, self-verification, and handoff artifacts across many context windows." walkinglabs_harness,54,Foundations,Harness design for long-running application development,https://www.anthropic.com/engineering/harness-design-long-running-apps,,,,,,grey,Anthropic follow-up focused on improving long-running app generation with better task state and evaluator design. walkinglabs_harness,55,Foundations,The Anatomy of an Agent Harness,https://blog.langchain.com/the-anatomy-of-an-agent-harness/,,,,,,grey,"LangChain's concise framing of an agent as model plus harness, with prompts, tools, middleware, orchestration, and runtime infrastructure." walkinglabs_harness,56,Foundations,Harness Engineering,https://martinfowler.com/articles/exploring-gen-ai/harness-engineering.html,,,,,,grey,"Thoughtworks' framing of harness work into context engineering, architectural constraints, and ""garbage collection"" against entropy." walkinglabs_harness,57,Foundations,Building effective agents,https://www.anthropic.com/engineering/building-effective-agents,,,,,,grey,"Anthropic's broader guide to workflows, agents, tools, and when structured systems outperform raw prompting." walkinglabs_harness,58,Foundations,Skill Issue: Harness Engineering for Coding Agents,https://www.humanlayer.dev/blog/skill-issue-harness-engineering-for-coding-agents,,,,,,grey,A practical argument that weak results from coding agents are often harness problems rather than model problems. walkinglabs_harness,59,Foundations,"Your Agent Needs a Harness, Not a Framework",https://www.inngest.com/blog/your-agent-needs-a-harness-not-a-framework,,,,,,grey,"Inngest's case for treating state, retries, traces, and concurrency as first-class infrastructure." walkinglabs_harness,60,Foundations,"Greenfield AI, Brownfield AI, and the Vibecode You Just Inherited",https://sawinyh.com/blog/greenfield-vs-brownfield-ai-codebases,,,,,,grey,"A three-way taxonomy of codebases agents encounter — agent-native greenfield, true legacy brownfield, and recently-vibecoded inheritance — with playbooks for installing layered CLAUDE.md rules, ratcheted pre-commit hooks, baselined lint violations, and feature-folder refactors so the codebase itself stops being the harness bottleneck." walkinglabs_harness,61,Foundations,"Harness Engineering for Language Agents: The Harness Layer as Control, Agency, and Runtime",https://www.preprints.org/manuscript/202603.1756,,,,,,grey,"A position paper that treats the harness layer as a first-class research object, proposes the control–agency–runtime (CAR) decomposition, and introduces HarnessCard for structured reporting of harness design and evaluation." walkinglabs_harness,62,Foundations,Many Hands Engineering,https://github.com/mseeks/many-hands-engineering/blob/main/many-hands-engineering.pdf,,,,,,repository,"A handbook framing the layer above the per-agent harness: how multiple harnessed agents share a commons, where decisions belong on a planned / emergent spectrum, and how human stewardship operates at a different cadence than agent execution. Treats harness engineering as a critical layer of ""terrain"" the framework sits on top of." walkinglabs_harness,63,Foundations,"Agent Frameworks, Runtimes, and Harnesses, Oh My!",https://blog.langchain.com/agent-frameworks-runtimes-and-harnesses-oh-my/,,,,,,grey,"LangChain's decomposition of what belongs in a framework, a runtime, and a harness." walkinglabs_harness,64,Foundations,"The Agent Loop: A Survey of Control Strategies, Skills, and Harnesses for LLM Agents",https://ssrn.com/abstract=7186738,,,,,,academic,"A survey that treats the agent loop, not the model, as the unit of analysis: loop paradigms, termination and step budgets, verification, context management and recovery, skills, and harnesses, with a chapter on separating harness effects from model effects in evaluation." walkinglabs_harness,65,Foundations,HARNESS-DB,https://github.com/harness-db/harness-db,,,,,,repository,"Open dataset of 1,256 LLM agent harnesses coded on 38 design dimensions (context, tools, loop, memory, verification, budgets, sandbox, observability), each value backed by a verbatim quote and a file:line@commit locator, so harness designs can be compared and checked against their source." walkinglabs_harness,71,"Context, Memory & Working State > Context Design & Delivery",Effective context engineering for AI agents,https://www.anthropic.com/engineering/effective-context-engineering-for-ai-agents,,,,,,grey,Anthropic's guidance on managing the context window as a working memory budget rather than a dumping ground. walkinglabs_harness,72,"Context, Memory & Working State > Context Design & Delivery",Context Engineering for AI Agents: Lessons from Building Manus,https://manus.im/blog/Context-Engineering-for-AI-Agents-Lessons-from-Building-Manus,,,,,,grey,"Manus' detailed playbook on KV-cache locality, tool masking, filesystem memory, and keeping useful failures in-context." walkinglabs_harness,73,"Context, Memory & Working State > Context Design & Delivery",Context Engineering for Coding Agents,https://martinfowler.com/articles/exploring-gen-ai/context-engineering-coding-agents.html,,,,,,grey,Thoughtworks guidance on shaping the task environment so coding agents can stay grounded and productive. walkinglabs_harness,74,"Context, Memory & Working State > Context Design & Delivery",Advanced Context Engineering for Coding Agents,https://www.humanlayer.dev/blog/advanced-context-engineering,,,,,,grey,HumanLayer patterns for reducing context drift and making coding sessions easier to resume. walkinglabs_harness,75,"Context, Memory & Working State > Context Design & Delivery",Context-Efficient Backpressure for Coding Agents,https://www.humanlayer.dev/blog/context-efficient-backpressure,,,,,,grey,HumanLayer's ideas for preventing agents from burning context on noisy or low-value work. walkinglabs_harness,76,"Context, Memory & Working State > Context Design & Delivery",OpenHands Context Condensensation for More Efficient AI Agents,https://openhands.dev/blog/openhands-context-condensensation-for-more-efficient-ai-agents,,,,,,grey,"OpenHands' design for bounded conversation memory that preserves goals, progress, critical files, and failing tests while keeping long-running coding sessions efficient." walkinglabs_harness,77,"Context, Memory & Working State > Context Design & Delivery",Writing a good CLAUDE.md,https://www.humanlayer.dev/blog/writing-a-good-claude-md,,,,,,grey,"A practical guide to creating durable, repo-local instructions that agents can repeatedly follow." walkinglabs_harness,78,"Context, Memory & Working State > Context Design & Delivery",Deterministic Context Routing,https://github.com/ai-erp-collab/deterministic-context-routing,,,,,,repository,"A context-management methodology that routes only the necessary-and-sufficient context to an agent through a deterministic chain (module registry → wiki → session state), cutting overread and context loss on large, under-documented multi-module codebases." walkinglabs_harness,79,"Context, Memory & Working State > Context Design & Delivery",DevProjex,https://github.com/Avazbek22/DevProjex,,,,,,repository,"GUI, TUI, CLI, and read-only MCP tooling for selecting and exporting structured codebase context with live preview, secret redaction enforced in MCP mode, Git scopes, token budgets, and syntax-aware compression." walkinglabs_harness,83,"Context, Memory & Working State > Memory & Knowledge Systems",wiki,https://github.com/plasma-ai/wiki,,,,,,repository,"Indexed Markdown knowledge bases that give agents incremental project context through deterministic indexes, scoped CLI access, and merge handling for parallel edits." walkinglabs_harness,84,"Context, Memory & Working State > Memory & Knowledge Systems",Data Olympus,https://github.com/knaisoma/data-olympus,,,,,,repository,"Git-native project knowledge base and MCP server with governed proposal-to-acceptance workflows, validity windows, supersession chains, and retrieval of in-force engineering guidance." walkinglabs_harness,86,"Context, Memory & Working State > Memory & Knowledge Systems",OpenViking,https://github.com/volcengine/OpenViking,,,,,,repository,"Context database that unifies agent memory, knowledge retrieval, and skills behind an MCP-accessible storage layer." walkinglabs_harness,87,"Context, Memory & Working State > Memory & Knowledge Systems",Hyperconsciousness,https://github.com/louis030195/hyperconsciousness,,,,,,repository,"Developer-alpha encrypted knowledge store that lets harnesses retrieve context and persist task handoffs through scoped, expiring MCP grants." walkinglabs_harness,88,"Context, Memory & Working State > Memory & Knowledge Systems",pentimento,https://github.com/kjiwa/pentimento,,,,,,repository,"Derives status and lineage for the plan files coding agents leave behind, so a later session or its operator can tell which plan finished, which replaced which, and find the decisions inside them." walkinglabs_harness,89,"Context, Memory & Working State > Memory & Knowledge Systems",Threadnote,https://github.com/threadnote/threadnote,,,,,,repository,"Local-first engineering memory for coding agents, with private session handoffs, reviewed knowledge tied to code citations, and optional Git-backed team sharing." walkinglabs_harness,95,"Constraints, Guardrails & Safe Autonomy > Tool Design & Execution Boundaries",Code execution with MCP: building more efficient agents,https://www.anthropic.com/engineering/code-execution-with-mcp,,,,,,grey,"Anthropic's approach to giving agents controlled execution power through explicit, inspectable tool boundaries." walkinglabs_harness,96,"Constraints, Guardrails & Safe Autonomy > Tool Design & Execution Boundaries",Writing effective tools for agents,https://www.anthropic.com/engineering/writing-tools-for-agents,,,,,,grey,Anthropic's guidance on tool interfaces that are easier for models to call correctly and safely. walkinglabs_harness,97,"Constraints, Guardrails & Safe Autonomy > Tool Design & Execution Boundaries",mcp-guardian,https://github.com/S1LV3RJ1NX/mcp-guardian,,,,,,repository,"MCP proxy for progressive tool discovery, scope-based allow/block lists, and audit logging, reducing startup context while preventing blocked tools from being discovered or called." walkinglabs_harness,98,"Constraints, Guardrails & Safe Autonomy > Tool Design & Execution Boundaries",solvi,https://github.com/solvi-ai/solvi,,,,,,repository,"Python runtime whose Guard checks each proposed tool call (tool in the catalog, typed arguments, values quoted from the conversation, policy rules) and allows it, denies it or escalates it to a person, keeping a replayable trace: a fail-closed execution boundary written as code." walkinglabs_harness,102,"Constraints, Guardrails & Safe Autonomy > Security, Authorization & Policy",Beyond permission prompts: making Claude Code more secure and autonomous,https://www.anthropic.com/engineering/claude-code-sandboxing,,,,,,grey,Anthropic on reducing approval friction without losing control through better sandboxing and policy design. walkinglabs_harness,103,"Constraints, Guardrails & Safe Autonomy > Security, Authorization & Policy",Mitigating Prompt Injection Attacks in Software Agents,https://openhands.dev/blog/mitigating-prompt-injection-attacks-in-software-agents,,,,,,grey,"OpenHands' practical guide to confirmation mode, analyzers, sandboxing, and hard policies for reducing prompt-injection risk in autonomous coding agents." walkinglabs_harness,104,"Constraints, Guardrails & Safe Autonomy > Security, Authorization & Policy",APort Agent Guardrails,https://github.com/aporthq/aport-agent-guardrails,,,,,,repository,"Deterministic pre-action authorization hooks for AI-agent tool calls, with adapters for Claude Code, Cursor, OpenClaw, LangChain, CrewAI, and related runtimes." walkinglabs_harness,105,"Constraints, Guardrails & Safe Autonomy > Security, Authorization & Policy",HOL Guard,https://github.com/hashgraph-online/hol-guard,,,,,,repository,"Open-source local policy checks and human approval for supported coding-agent actions including shell commands, file access, package installs, and MCP tool calls; coverage depends on the agent integration." walkinglabs_harness,106,"Constraints, Guardrails & Safe Autonomy > Security, Authorization & Policy",Lurkr,https://github.com/agentveil-protocol/lurkr,,,,,,repository,"Static scanner that runs in CI before deploy to surface AI-agent capability risks, including shadow capabilities, credentials into LLM context, eval/subprocess in @tool, direct prompt interpolation, and unverified MCP endpoints." walkinglabs_harness,107,"Constraints, Guardrails & Safe Autonomy > Security, Authorization & Policy",HEAAL,https://github.com/hyun06000/AIL,,,,,,repository,Grammar-enforced safety constraints for AI agents via AIL (AI-Intent Language). walkinglabs_harness,108,"Constraints, Guardrails & Safe Autonomy > Security, Authorization & Policy",Odai,https://github.com/orziz/odai,,,,,,repository,"Cross-host governance layer that binds intent, evidence, authorization, risk, and acceptance to agent execution, with a portable Skill, inspectable DSH runtime, and published version-scoped paired evaluations." walkinglabs_harness,109,"Constraints, Guardrails & Safe Autonomy > Security, Authorization & Policy",sofagent,https://github.com/KongFangXun/sofagent,,,,,,repository,"Commit-time audit harness for AI coding agents: 28 deterministic, zero-token audit rules over Git diffs (secrets, out-of-scope edits, blind modifications, prompt-injection traces), an HMAC-chained tamper-evident audit history, and a DSH plugin family; installable as a GitHub Action or MCP server." walkinglabs_harness,110,"Constraints, Guardrails & Safe Autonomy > Security, Authorization & Policy",Orca AI Incident Archive,https://github.com/Continuum-AI-Corp/Orca-AI-Incident-Archive,,,,,,repository,"Sourced database of real-world AI agent security incidents, useful as threat-model input for harness boundaries: sandbox escapes, destructive agent actions, prompt injection via repo content and agent-framework CVEs, each flagged for confirmed harm vs. demonstration only." walkinglabs_harness,114,"Constraints, Guardrails & Safe Autonomy > Operational Autonomy & Reliability","Spend rails for autonomous agents: a number, not a vibe",https://joeyycli.github.io/agent-ops-kit-guide/docs/spend-rails-for-autonomous-agents.html,,,,,,grey,"A concrete pattern for giving a scheduled agent real purchasing authority without real risk: a per-transaction autonomy line, an escalation threshold, a hard lifetime ceiling, and a transaction ledger as the actual enforcement mechanism instead of relying on the model's judgment." walkinglabs_harness,115,"Constraints, Guardrails & Safe Autonomy > Operational Autonomy & Reliability",Distributed retry patterns: bounding blast radius across a fleet,https://loopandretry.github.io/posts/fleet-retry-patterns/,,,,,,grey,"Practitioner guidance on concurrency bounds, decorrelated backoff, circuit breakers, and idempotency for preventing agent retry storms." walkinglabs_harness,116,"Constraints, Guardrails & Safe Autonomy > Operational Autonomy & Reliability",How to add a circuit breaker to an AI agent loop,https://www.robatdasorvi.com/chapters/ai-agents/why-every-agentic-loop-needs-a-circuit-breaker-and-how-i-built-one,,,,,,grey,"Post-incident write-up from a scheduled agent whose lockout check fired while the loop ran 3.5 more days, then a second incident where detection missed entirely; argues for a global action integral, a duration bound, a process exit with a human-cleared sentinel, and tripping on missing success rather than only on known failure screens." walkinglabs_harness,117,"Constraints, Guardrails & Safe Autonomy > Operational Autonomy & Reliability",What breaks when you run twenty agents at once,https://jedarden.com/notes/what-breaks-at-twenty-agents/,,,,,,grey,"Failure taxonomy from a year of unattended coding-agent fleets: eight ways a fleet dies, none of them the model, and the operational discipline they add up to." walkinglabs_harness,118,"Constraints, Guardrails & Safe Autonomy > Operational Autonomy & Reliability",Decomposition is concurrency control,https://jedarden.com/guides/workflow/#s06-decomposition-is-concurrency-control,,,,,,grey,"Cutting work into tasks is also choosing the concurrency: a design-time file partition (""Owns"" lines) serialises overlapping tasks through dependency edges instead of runtime negotiation." walkinglabs_harness,124,"Specs, Agent Files & Workflow Design > Instruction Files & Formats",AGENTS.md,https://github.com/agentsmd/agents.md,,,,,,repository,A lightweight open format for repo-local instructions that tell agents how to work inside a codebase. walkinglabs_harness,125,"Specs, Agent Files & Workflow Design > Instruction Files & Formats",agent.md,https://github.com/agentmd/agent.md,,,,,,repository,A related standardization effort for machine-readable agent instructions across projects and tools. walkinglabs_harness,129,"Specs, Agent Files & Workflow Design > Spec-Driven Development",GitHub Spec Kit,https://github.com/github/spec-kit,,,,,,repository,"GitHub's toolkit for spec-driven development, useful when you want agents to execute against explicit product and engineering specs." walkinglabs_harness,130,"Specs, Agent Files & Workflow Design > Spec-Driven Development",Context Repository-Driven Development (CRDD),https://github.com/qual-lab/CRDD,,,,,,repository,"](https://github.com/qual-lab/CRDD) - A repository-centered methodology for preserving product intent, decisions, specifications, evidence, and traceability as durable context while keeping approval authority with humans." walkinglabs_harness,131,"Specs, Agent Files & Workflow Design > Spec-Driven Development",Kibi,https://github.com/Looted/kibi,,,,,,repository,"Repo-local requirements linked to code and tests, with checks that help coding agents stay aligned with project intent." walkinglabs_harness,132,"Specs, Agent Files & Workflow Design > Spec-Driven Development","Understanding Spec-Driven-Development: Kiro, spec-kit, and Tessl",https://martinfowler.com/articles/exploring-gen-ai/sdd-3-tools.html,,,,,,grey,Thoughtworks on why strong specs make AI-assisted software delivery more dependable. walkinglabs_harness,133,"Specs, Agent Files & Workflow Design > Spec-Driven Development",Archcore,https://github.com/archcore-ai/archcore,,,,,,repository,"Spec-driven development and context engineering for Claude Code, Cursor, Codex, and GitHub Copilot — backed by project context in Git." walkinglabs_harness,137,"Specs, Agent Files & Workflow Design > Operating Principles & Human Oversight",12 Factor Agents,https://www.humanlayer.dev/blog/12-factor-agents,,,,,,grey,"HumanLayer's operating principles for production agents, including explicit prompts, state ownership, and clean pause-resume behavior." walkinglabs_harness,138,"Specs, Agent Files & Workflow Design > Operating Principles & Human Oversight",12-Factor AgentOps,https://www.12factoragentops.com/,,,,,,grey,"An operations-oriented companion focused on context discipline, validation, and reproducible agent workflows." walkinglabs_harness,139,"Specs, Agent Files & Workflow Design > Operating Principles & Human Oversight",Anchoring AI to a reference application,https://martinfowler.com/articles/exploring-gen-ai/anchoring-to-reference.html,,,,,,grey,Thoughtworks on constraining agents with concrete exemplars so they produce more consistent output. walkinglabs_harness,140,"Specs, Agent Files & Workflow Design > Operating Principles & Human Oversight",Humans and Agents in Software Engineering Loops,https://martinfowler.com/articles/exploring-gen-ai/humans-and-agents.html,,,,,,grey,A clear mental model for where humans should strengthen the harness instead of micromanaging every artifact. walkinglabs_harness,141,"Specs, Agent Files & Workflow Design > Operating Principles & Human Oversight",Claude Code: Best practices for agentic coding,https://code.claude.com/docs,,,,,,grey,"Anthropic's practical recommendations for repo structure, checkpoints, validation, and delegation in agentic coding workflows." walkinglabs_harness,147,Evals & Observability > Evaluation Design,Testing Agent Skills Systematically with Evals,https://developers.openai.com/blog/eval-skills/,,,,,,grey,OpenAI's concrete guide to turning agent traces into repeatable evals with JSONL logs and deterministic checks. walkinglabs_harness,148,Evals & Observability > Evaluation Design,How to Evaluate Agent Skills (And Why You Should),https://openhands.dev/blog/evaluating-agent-skills,,,,,,grey,"](https://openhands.dev/blog/evaluating-agent-skills) - OpenHands' hands-on playbook for measuring whether a skill actually helps using bounded tasks, deterministic verifiers, no-skill baselines, and trace review." walkinglabs_harness,149,Evals & Observability > Evaluation Design,Agent evals,https://platform.openai.com/docs/guides/agent-evals,,,,,,grey,OpenAI's product guide for measuring agent quality with reproducible task-level and workflow-level evaluations. walkinglabs_harness,150,Evals & Observability > Evaluation Design,Evaluation best practices,https://platform.openai.com/docs/guides/evaluation-best-practices,,,,,,grey,OpenAI's general guide to building eval suites that match real-world distributions and catch regressions early. walkinglabs_harness,151,Evals & Observability > Evaluation Design,Trace grading,https://platform.openai.com/docs/guides/trace-grading,,,,,,grey,"OpenAI documentation on grading agent traces directly, which is especially helpful for long multi-step tasks." walkinglabs_harness,152,Evals & Observability > Evaluation Design,Inspect AI,https://inspect.aisi.org.uk/,,,,,,grey,"UK AISI's open-source evaluation framework with solver, scorer, sandboxing, tool-use, MCP, and log-viewer primitives for building reproducible agent eval harnesses." walkinglabs_harness,153,Evals & Observability > Evaluation Design,Demystifying Evals for AI Agents,https://www.anthropic.com/engineering/demystifying-evals-for-ai-agents,,,,,,grey,Anthropic's guidance on what to measure when agents have many possible trajectories to success or failure. walkinglabs_harness,154,Evals & Observability > Evaluation Design,Evaluating Deep Agents: Our Learnings,https://blog.langchain.com/evaluating-deep-agents-our-learnings/,,,,,,grey,"LangChain's practical breakdown of single-step, full-run, and multi-turn eval design for stateful agents." walkinglabs_harness,156,Evals & Observability > Evaluation Design,Training Agents to Evolve with Their Harness: TaoLive Digital Avatar Agent Technical Report,https://arxiv.org/abs/2608.15763,2608.15763,,2026,8,arxiv_id,academic,"Technical report on training agents for changing harnesses, with held-out prompt and Skill edit evaluations that keep model weights fixed and document developer review." walkinglabs_harness,160,Evals & Observability > Verification & Quality Gates,Agent Role Contracts (ARC),https://github.com/suirindo/agent-role-contracts,,,,,,repository,"](https://github.com/suirindo/agent-role-contracts) - Offline Node.js preflight checks for declared coding-agent write scopes, separate read-only review roles, and handoffs. Produces deterministic failures for inconsistent declarations; runtime enforcement stays with the host." walkinglabs_harness,161,Evals & Observability > Verification & Quality Gates,Assessing internal quality while coding with an agent,https://martinfowler.com/articles/exploring-gen-ai/ccmenu-quality.html,,,,,,grey,Thoughtworks on moving quality checks into the loop instead of relying on after-the-fact manual review. walkinglabs_harness,162,Evals & Observability > Verification & Quality Gates,Better Harness,https://github.com/QoderAI/better-harness,,,,,,repository,"Reviewer for coding-agent workflows that turns repository and session evidence into prioritized, verifiable harness improvements while keeping unobserved behavior explicit." walkinglabs_harness,163,Evals & Observability > Verification & Quality Gates,Learning to Verify AI-Generated Code,https://openhands.dev/blog/20260305-learning-to-verify-ai-generated-code,,,,,,grey,"OpenHands' overview of a layered verification stack using trajectory critics trained on production traces for reranking, early stopping, and review-time quality control." walkinglabs_harness,164,Evals & Observability > Verification & Quality Gates,Improving Deep Agents with harness engineering,https://blog.langchain.com/improving-deep-agents-with-harness-engineering/,,,,,,grey,LangChain's evidence that harness changes alone can significantly improve benchmark performance. walkinglabs_harness,165,Evals & Observability > Verification & Quality Gates,self-auditing-agent,https://github.com/simin-yuan/self-auditing-agent,,,,,,repository,"A public audit log of one agent's real task in which every conclusion carries a runnable command and the record keeps the agent's own four bugs, one false positive, and one false discovery it caught itself: a concrete reference for making an agent's claims checkable by a third party and its failures part of the permanent record." walkinglabs_harness,166,Evals & Observability > Verification & Quality Gates,RunLedger,https://github.com/runledger/Runledger,,,,,,repository,"CI harness that records an agent's tool calls once and replays them in CI, so a PR fails when tool order or a JSON-schema contract breaks, a tool-call budget is exceeded, or a baseline regresses; emits JUnit and HTML reports. This makes harness verification a merge gate." walkinglabs_harness,167,Evals & Observability > Verification & Quality Gates,Don't let the agent grade its own homework,https://jedarden.com/notes/dont-let-the-agent-grade-its-own-homework/,,,,,,grey,"An agent reporting ""done"" is the actor grading its own work; the cheap guard is an independent observer — a second agent, a separate tool, or the live artifact — spawned interactively and as a validation gate in a fleet." walkinglabs_harness,171,"Evals & Observability > Telemetry, Tracing & Performance",OpenTelemetry Semantic Conventions for Generative AI Systems,https://opentelemetry.io/docs/specs/semconv/gen-ai/,,,,,,grey,"Standard span, metric, event, and attribute conventions for instrumenting LLM and agent workflows so harness traces stay portable across observability backends." walkinglabs_harness,172,"Evals & Observability > Telemetry, Tracing & Performance",AgentOps,https://github.com/AgentOps-AI/agentops,,,,,,repository,"Open-source Python SDK for agent monitoring, session replay, cost tracking, benchmarking, and tracing across common LLM and agent frameworks." walkinglabs_harness,173,"Evals & Observability > Telemetry, Tracing & Performance",agenttrace,https://github.com/luoyuctl/agenttrace,,,,,,repository,"Local-first TUI/CLI for auditing AI coding-agent session traces, health gates, cost spikes, tool failures, latency gaps, and attempt-to-attempt diffs." walkinglabs_harness,174,"Evals & Observability > Telemetry, Tracing & Performance",flameox,https://github.com/morluto/flameox,,,,,,repository,"Profiling and optimization toolkit for agents: bounded CLI and MCP workflows capture traces, preserve native evidence, and compare experiments behind runtime conclusions." walkinglabs_harness,176,"Evals & Observability > Telemetry, Tracing & Performance",traceAI,https://github.com/future-agi/traceAI,,,,,,repository,"Apache-2.0 OpenTelemetry instrumentation for 50+ AI frameworks across Python, TypeScript, Java and C# that records LLM calls, tool calls, retrieval and agent steps as standard spans, so harness traces stay portable to any OTel backend." walkinglabs_harness,177,"Evals & Observability > Telemetry, Tracing & Performance",Quantifying infrastructure noise in agentic coding evals,https://www.anthropic.com/engineering/infrastructure-noise,,,,,,grey,Anthropic on how runtime configuration can move coding benchmark scores by more than many leaderboard gaps. walkinglabs_harness,185,Benchmarks > Coding & Terminal Agents,EvoClaw: Evaluating AI Agents on Continuous Software Evolution,https://openhands.dev/blog/evoclaw-benchmark,,,,,,grey,"A benchmark write-up on evaluating agents across dependent milestone sequences from real repository history, surfacing regression accumulation and long-horizon precision loss." walkinglabs_harness,186,Benchmarks > Coding & Terminal Agents,Introducing Terminal-Bench 2.0 and Harbor,https://www.tbench.ai/news/announcement-2-0,,,,,,grey,"The Terminal-Bench 2.0 announcement, useful for understanding the harder tasks and generalized evaluation harness behind Harbor." walkinglabs_harness,187,Benchmarks > Coding & Terminal Agents,LeetCode-Hard Gym,https://github.com/GammaTauAI/leetcode-hard-gym,,,,,,repository,"An RL environment interface to LeetCode's submission server for evaluating codegen agents, giving harnesses direct access to execution-based feedback on hard algorithmic problems." walkinglabs_harness,188,Benchmarks > Coding & Terminal Agents,SEC-bench,https://github.com/SEC-bench/SEC-bench,,,,,,repository,"A benchmark for evaluating LLM agents on real-world software security tasks including vulnerability reproduction and patching, stressing harness design around code execution, containerized environments, and security-aware tooling." walkinglabs_harness,189,Benchmarks > Coding & Terminal Agents,SWE-bench Verified,https://www.swebench.com/,,,,,,grey,"A strong benchmark for software engineering agents working against real GitHub issues and tests, which makes harness choices around retrieval, patching, and validation highly visible." walkinglabs_harness,190,Benchmarks > Coding & Terminal Agents,Terminal-Bench,https://www.tbench.ai/,,,,,,grey,"A benchmark suite for terminal-native agents operating in shells, filesystems, and verification-heavy environments, which is especially useful for comparing coding-agent harnesses." walkinglabs_harness,191,Benchmarks > Coding & Terminal Agents,LoopArena,https://github.com/AMAP-ML/LoopArena,,,,,,repository,"Open benchmark for evaluating models as runtime Controllers around a separate, fixed coding Worker. [Paper](https://huggingface.co/papers/2608.28281) · [ModelScope](https://www.modelscope.cn/papers/2608.28281)" walkinglabs_harness,195,"Benchmarks > Web, GUI & Computer Use",AgentStudio,https://github.com/SkyworkAI/agent-studio,,,,,,repository,"An integrated benchmark suite with realistic environments and comprehensive toolkits for evaluating virtual agents on real computer software, useful for measuring harness depth against a broad task surface." walkinglabs_harness,196,"Benchmarks > Web, GUI & Computer Use",AssistantBench,https://github.com/oriyor/AssistantBench,,,,,,repository,"A benchmark that evaluates web agents on realistic, time-consuming research tasks requiring multi-step tool use and information synthesis, making it a good proxy for harness quality in long-horizon web scenarios." walkinglabs_harness,197,"Benchmarks > Web, GUI & Computer Use",BrowseComp,https://www.kaggle.com/benchmarks/openai/browsecomp,,,,,,grey,"A benchmark that evaluates AI agents on locating hard-to-find information, stressing search strategy, context management, and retrieval harness design under difficult conditions." walkinglabs_harness,198,"Benchmarks > Web, GUI & Computer Use",BrowserGym Leaderboard,https://huggingface.co/spaces/ServiceNow/browsergym-leaderboard,,,,,,repository,"A gym environment and leaderboard for evaluating LLMs, VLMs, and agents on web navigation tasks, offering a reproducible framework for comparing harnesses across multiple web benchmarks in one place." walkinglabs_harness,199,"Benchmarks > Web, GUI & Computer Use",ClawBench: Can AI Agents Complete Everyday Online Tasks?,https://huggingface.co/papers/2604.08523,,,,,,repository,"A browser-agent benchmark of 153 everyday web tasks across 144 live production sites in 15 categories, using a lightweight interception layer that captures and blocks only the final submission request so agents can be scored end-to-end on real websites without real-world side effects." walkinglabs_harness,200,"Benchmarks > Web, GUI & Computer Use",Computer Agent Arena,https://github.com/xlang-ai/computer-agent-arena,,,,,,repository,"An open evaluation platform where users compare LLM/VLM-based agents on real-world computer tasks ranging from general computer use to coding, data analysis, and video editing, surfacing harness differences across a wide task surface." walkinglabs_harness,201,"Benchmarks > Web, GUI & Computer Use",OSWorld,https://os-world.github.io/,,,,,,grey,"A real computer-use benchmark with 369 tasks across Ubuntu, Windows, and macOS, complete with initial-state setup and execution-based evaluators, making it excellent for testing desktop and multimodal harnesses." walkinglabs_harness,202,"Benchmarks > Web, GUI & Computer Use",OSWorld-MCP,https://osworld-mcp.github.io,,,,,,grey,"An extension of OSWorld that evaluates AI agents on real-world computer tasks using the Model Context Protocol, making it useful for comparing MCP-enabled harnesses on a realistic desktop task suite." walkinglabs_harness,204,"Benchmarks > Web, GUI & Computer Use",VisualWebArena,https://jykoh.com/vwa,,,,,,grey,"A benchmark for multimodal web agents on realistic visually grounded tasks, extending WebArena with image and screenshot inputs that stress harness support for visual context in browser environments." walkinglabs_harness,205,"Benchmarks > Web, GUI & Computer Use",WebArena,https://webarena.dev/,,,,,,grey,"A standalone, self-hostable web environment for evaluating autonomous agents on realistic tasks, making it a reproducible baseline for comparing web-facing harness designs." walkinglabs_harness,206,"Benchmarks > Web, GUI & Computer Use",WebArena-Verified,https://github.com/ServiceNow/webarena-verified,,,,,,repository,"A verified web-agent benchmark with curated tasks and deterministic evaluators over agent responses and captured network traces, making it a good fit for measuring web-facing harnesses." walkinglabs_harness,207,"Benchmarks > Web, GUI & Computer Use",WebsiteBench,https://github.com/TIGER-AI-Lab/WebsiteBench,,,,,,repository,Website reconstruction tasks with native Harbor configurations and a separate no-network verifier for checking coding-agent implementations. walkinglabs_harness,208,"Benchmarks > Web, GUI & Computer Use",WorkArena,https://github.com/ServiceNow/WorkArena,,,,,,repository,"A benchmark for browser agents on common knowledge-work tasks, useful for comparing harnesses on realistic enterprise-style web workflows instead of toy browser tasks." walkinglabs_harness,212,"Benchmarks > Tools, APIs & MCP",AgentBench,https://github.com/THUDM/AgentBench,,,,,,repository,"A cross-environment benchmark spanning OS, databases, knowledge graphs, web browsing, and more, useful for seeing whether a harness generalizes beyond one narrow task loop." walkinglabs_harness,213,"Benchmarks > Tools, APIs & MCP",AppWorld,https://appworld.dev/,,,,,,grey,"A controllable world of apps and people for benchmarking interactive coding agents, with state-based and execution-based unit tests that surface harness quality around planning, code generation, and collateral-damage control." walkinglabs_harness,215,"Benchmarks > Tools, APIs & MCP",MCP Bench,https://github.com/modelscope/MCPBench,,,,,,repository,"A benchmark for evaluating AI models on MCP server interactions, measuring tool accuracy, latency, and token use across server types, which directly reflects harness design choices around MCP integration." walkinglabs_harness,216,"Benchmarks > Tools, APIs & MCP",MCP Universe,https://mcp-universe.github.io/,,,,,,grey,"A leaderboard comparing AI model performance on MCP tasks, tracking how different models and harness configurations handle tool-augmented agent workflows." walkinglabs_harness,217,"Benchmarks > Tools, APIs & MCP",MCPMark,https://github.com/eval-sys/mcpmark,,,,,,repository,"A stress-testing benchmark for model and agent capabilities in real-world MCP tasks across tools like Notion, GitHub, and PostgreSQL, making harness MCP integration quality directly measurable." walkinglabs_harness,218,"Benchmarks > Tools, APIs & MCP",τ-Bench,https://github.com/sierra-research/tau-bench,,,,,,repository,"A benchmark that emulates dynamic conversations between a simulated user and a language agent equipped with domain-specific API tools and policy guidelines, making it useful for evaluating harnesses built around structured tool use and policy enforcement." walkinglabs_harness,219,"Benchmarks > Tools, APIs & MCP",tau2-bench,https://github.com/sierra-research/tau2-bench,,,,,,repository,"A benchmark for realistic, multi-step agent tasks where success depends on tool use and execution quality rather than a single-shot answer." walkinglabs_harness,220,"Benchmarks > Tools, APIs & MCP",TravelPlanner,https://github.com/OSU-NLP-Group/TravelPlanner,,,,,,repository,"A benchmark for evaluating LLM agents on tool use and complex planning within multiple constraints, revealing how harness design handles multi-constraint satisfaction and long-horizon planning." walkinglabs_harness,224,"Benchmarks > Multi-Agent, General & Interactive",Agent Arena,https://github.com/ShishirPatil/gorilla/tree/main/agent-arena,,,,,,repository,"Source code for the Gorilla project’s agent comparison platform, including its evaluation and leaderboard frontend; the original hosted leaderboard is currently unavailable." walkinglabs_harness,225,"Benchmarks > Multi-Agent, General & Interactive",AgentBoard,https://github.com/HKUST-NLP/AgentBoard,,,,,,repository,"A benchmark for multi-turn LLM agents complemented by an analytical evaluation board for assessing model performance beyond final success rates, making partial-progress and trajectory quality visible." walkinglabs_harness,226,"Benchmarks > Multi-Agent, General & Interactive",CharacterEval,https://github.com/morecry/CharacterEval,,,,,,repository,"A benchmark for evaluating role-playing conversational agents using multi-turn dialogues and character profiles, with metrics across four dimensions including character fidelity and conversational coherence." walkinglabs_harness,227,"Benchmarks > Multi-Agent, General & Interactive",ClawBench,https://github.com/claw-bench/claw-bench,,,,,,repository,"A benchmark that evaluates AI agents across search, reasoning, coding, safety, and multi-turn conversation tasks, covering the breadth of harness demands in a single suite." walkinglabs_harness,228,"Benchmarks > Multi-Agent, General & Interactive",GAIA,https://huggingface.co/datasets/gaia-benchmark/GAIA,,,,,,repository,"A benchmark for general AI assistants that is often used to compare harness-level choices around tools, planning, verification, and long-horizon autonomy." walkinglabs_harness,229,"Benchmarks > Multi-Agent, General & Interactive",Galileo Agent Leaderboard,https://huggingface.co/spaces/galileo-ai/agent-leaderboard,,,,,,repository,"An open evaluation platform tracking LLM agents on task completion and tool calling across business domains, useful for comparing harness quality in enterprise-grade agentic scenarios." walkinglabs_harness,230,"Benchmarks > Multi-Agent, General & Interactive",HAL: Holistic Agent Leaderboard,https://hal.cs.princeton.edu/,,,,,,grey,"A benchmark and leaderboard for agent systems with attention to reliability, cost, and broad task coverage, making it useful for comparing end-to-end harness behavior." walkinglabs_harness,231,"Benchmarks > Multi-Agent, General & Interactive",LLM Colosseum Leaderboard,https://github.com/OpenGenerativeAI/llm-colosseum,,,,,,repository,"A platform that evaluates LLMs by having them fight in Street Fighter III, testing speed, adaptability, and real-time decision-making as proxies for harness responsiveness under tight latency constraints." walkinglabs_harness,232,"Benchmarks > Multi-Agent, General & Interactive",MAgIC,https://zhiyuanhubj.github.io/MAgIC/,,,,,,grey,"A benchmark measuring cognition, adaptability, rationality, and collaboration of LLMs in multi-agent systems, useful for evaluating how harnesses coordinate agent interactions and shared state." walkinglabs_harness,233,"Benchmarks > Multi-Agent, General & Interactive",WildClawBench,https://github.com/InternLM/WildClawBench,,,,,,repository,"An in-the-wild benchmark running agents inside a live OpenClaw environment on 60 original tasks including multimodal, long-horizon, and safety-critical scenarios, making harness robustness under real-world conditions directly visible." walkinglabs_harness,237,"Benchmarks > Safety, Robustness & Economic Agency",ClawWork,https://github.com/HKUDS/ClawWork,,,,,,repository,"A real-world economic benchmark where AI agents complete professional tasks spanning 44 occupations, earning income while managing token costs and economic solvency, making it a direct test of harness efficiency under resource constraints." walkinglabs_harness,243,"Runtimes, Harnesses & Reference Implementations > Runtime Foundations & Control Layers",Building agents with the Claude Agent SDK,https://claude.com/blog/building-agents-with-the-claude-agent-sdk,,,,,,grey,"Anthropic's guide to a production-oriented agent SDK with sessions, tools, and orchestration support." walkinglabs_harness,244,"Runtimes, Harnesses & Reference Implementations > Runtime Foundations & Control Layers",AgentKit,https://github.com/inngest/agent-kit,,,,,,repository,"Inngest's TypeScript toolkit for building durable, workflow-aware agents on top of event-driven infrastructure." walkinglabs_harness,245,"Runtimes, Harnesses & Reference Implementations > Runtime Foundations & Control Layers",SandBase Harness,https://github.com/sandbaseai/sandbase-harness,,,,,,repository,"Apache-2.0 agent runtime with persistent sessions, governed MCP tools, credential handling, audit and replay, and interchangeable local or sandboxed execution backends." walkinglabs_harness,246,"Runtimes, Harnesses & Reference Implementations > Runtime Foundations & Control Layers",DSH Studio,https://github.com/Moresyl/dsh-studio,,,,,,repository,"Cross-platform desktop host for DeepSeek Harness with health probes, restart backoff, collision-free ports, and whole-process-tree cleanup." walkinglabs_harness,247,"Runtimes, Harnesses & Reference Implementations > Runtime Foundations & Control Layers",BitRouter,https://github.com/bitrouter/bitrouter,,,,,,repository,"Apache-2.0 model router with cross-protocol routing, MCP gateway, guardrails, observability, virtual keys, and multi-account failover." walkinglabs_harness,248,"Runtimes, Harnesses & Reference Implementations > Runtime Foundations & Control Layers",rust-norion,https://github.com/yanghao1143/rust-norion,,,,,,repository,"GPL-3.0 Rust inference-control prototype exploring runtime boundaries, governed memory and replay, evidence-based writer gates, audit traces, and rollback for self-evolving agent systems." walkinglabs_harness,249,"Runtimes, Harnesses & Reference Implementations > Runtime Foundations & Control Layers",Weave Router,https://github.com/weave-os/router,,,,,,repository,"Apache-2.0 drop-in proxy for the Anthropic, OpenAI and Gemini APIs that sits under harnesses like Claude Code, Codex, opencode and pi and picks the model for each request with an on-box embedder; self-hostable, BYOK, OTLP traces." walkinglabs_harness,250,"Runtimes, Harnesses & Reference Implementations > Runtime Foundations & Control Layers",theloop,https://github.com/paveler/theloop,,,,,,repository,"Apache-2.0 agent-loop engine behind a stateless HTTP API: the harness owns history, approvals, and billing, while write tools return deferred intents, so mutations wait for the harness to approve them." walkinglabs_harness,251,"Runtimes, Harnesses & Reference Implementations > Runtime Foundations & Control Layers",CUGA,https://github.com/cuga-project/cuga-agent,,,,,,repository,"IBM Research's Apache-2.0 agent harness on LangGraph: a planner and code-act executor keep intermediate results in variables instead of the prompt, and policies (intent guards, playbooks, tool approval) gate actions with human-in-the-loop approval across API and browser tools." walkinglabs_harness,255,"Runtimes, Harnesses & Reference Implementations > Sandboxes & Execution Infrastructure",SWE-ReX,https://github.com/SWE-agent/SWE-ReX,,,,,,repository,"Sandboxed code execution infrastructure for AI agents, useful when harness work starts to merge into execution runtime design." walkinglabs_harness,256,"Runtimes, Harnesses & Reference Implementations > Sandboxes & Execution Infrastructure",Harbor,https://github.com/harbor-framework/harbor,,,,,,repository,"A generalized harness for evaluating and improving agents at scale, released alongside Terminal-Bench 2.0." walkinglabs_harness,261,"Runtimes, Harnesses & Reference Implementations > Coding-Agent Harnesses",TeaQL Agent Kit,https://github.com/teaql/teaql-agent-kit,,,,,,repository,"Model-mediated coding-agent workflow for database-backed applications, combining deterministic domain-model evaluation and repair, typed code generation, and evidence-backed build/test verification." walkinglabs_harness,262,"Runtimes, Harnesses & Reference Implementations > Coding-Agent Harnesses",deepagents,https://github.com/langchain-ai/deepagents,,,,,,repository,"LangChain's open-source project for building deeper, longer-running agents with middleware and harness patterns." walkinglabs_harness,263,"Runtimes, Harnesses & Reference Implementations > Coding-Agent Harnesses",SWE-agent,https://github.com/SWE-agent/SWE-agent,,,,,,repository,"A mature research coding agent that makes the harness, prompt, tools, and environment design directly inspectable." walkinglabs_harness,264,"Runtimes, Harnesses & Reference Implementations > Coding-Agent Harnesses",Citadel,https://github.com/SethGammon/Citadel,,,,,,repository,"A harness for Claude Code and OpenAI Codex with isolated worktrees, multi-agent coordination, and persisted memory and campaign state." walkinglabs_harness,265,"Runtimes, Harnesses & Reference Implementations > Coding-Agent Harnesses",Harness Evolver,https://github.com/raphaelchristi/harness-evolver,,,,,,repository,"Claude Code plugin that autonomously evolves LLM agent harnesses using multi-agent proposers, LangSmith-backed evaluation, and Git worktree isolation. Based on Meta-Harness (Lee et al., 2026)." walkinglabs_harness,266,"Runtimes, Harnesses & Reference Implementations > Coding-Agent Harnesses",Ralph Wiggum as a Software Engineer,https://ghuntley.com/ralph/,,,,,,grey,"Geoffrey Huntley's write-up of ""Ralph,"" a minimalist while :; do cat PROMPT.md | claude-code; done harness pattern that uses single-task loops, deterministic prompt stacking, and bounded subagent parallelism to drive long-running autonomous coding." walkinglabs_harness,267,"Runtimes, Harnesses & Reference Implementations > Coding-Agent Harnesses",RailWarden,https://github.com/advaith-1212/railwarden,,,,,,repository,"Deterministic control plane for multi-agent software work with dependency-aware packages, isolated worktrees, durable validation evidence, recovery checkpoints, and integration gates." walkinglabs_harness,268,"Runtimes, Harnesses & Reference Implementations > Coding-Agent Harnesses",OpenCode Agent Orchestration Kit,https://github.com/jcarlosrodicio/opencode-agent-orchestration-kit,,,,,,repository,"Reproducible OpenCode harness with role-based agents, explicit handoffs, repo-local skills, safe installation, and mechanical contract validation." walkinglabs_harness,269,"Runtimes, Harnesses & Reference Implementations > Coding-Agent Harnesses",LoopTroop,https://github.com/looptroop-ai/LoopTroop,,,,,,repository,"Local-first GUI harness for long-running coding work with multi-model planning, isolated worktrees, and fresh-context recovery loops." walkinglabs_harness,270,"Runtimes, Harnesses & Reference Implementations > Coding-Agent Harnesses",Agent AFK,https://github.com/griffinwork40/agent-afk,,,,,,repository,"Headless coding-agent harness for asynchronous runs with explicit terminal states, editable lifecycle hooks, permission gates, model routing, and append-only traces." walkinglabs_harness,271,"Runtimes, Harnesses & Reference Implementations > Coding-Agent Harnesses",completely,https://github.com/23ag1/completely,,,,,,repository,"Claude Code plugin harness with a default-fail evaluator, deterministic write and close gates, orphan recovery, and parallel-worker integration checks." walkinglabs_harness,272,"Runtimes, Harnesses & Reference Implementations > Coding-Agent Harnesses",forge-harness,https://github.com/chrono-meta/forge-harness,,,,,,repository,"Claude Code plugin for adversarial validation, source-grounding audits, session-learning capture, and pre-deployment transfer simulation." walkinglabs_harness,273,"Runtimes, Harnesses & Reference Implementations > Coding-Agent Harnesses",YYLO,https://github.com/yylo-dev/yylo,,,,,,repository,"Command-line harness that runs coding agents (Claude Code, Codex, Gemini CLI) across isolated task worktrees, with a Kanban task lifecycle, review gates, and typed merge and release-readiness boundaries." walkinglabs_harness,275,"Runtimes, Harnesses & Reference Implementations > Coding-Agent Harnesses",Orbi,https://github.com/orbi-build/orbi,,,,,,repository,"Issue-driven harness on top of the Pi coding agent: each labelled GitHub issue runs in an isolated worktree, a separate review session must pass the exact commit before it merges, and all run state lives in issues, labels and PRs so a restart resumes the same run." walkinglabs_harness,276,"Runtimes, Harnesses & Reference Implementations > Coding-Agent Harnesses",Frink,https://github.com/benord-labs/frink,,,,,,repository,"Desktop-first, open-source (MIT) application for running parallel agents, and multi-step custom workflows triggered by schedules, webhooks or events from tools like Linear and Sentry, across your already-used providers." walkinglabs_harness,277,"Runtimes, Harnesses & Reference Implementations > Coding-Agent Harnesses",Zaru,https://github.com/100monkeys-ai/zaru-cli,,,,,,repository,"Apache-2.0 terminal harness in Rust that writes every model reply, tool call, permission decision and tool result to a plain-text JSONL transcript, and runs project-declared validator commands in a loop, feeding failures back to the model until they pass or an iteration limit is reached; pre-alpha, Linux only, built from source." walkinglabs_harness,278,"Runtimes, Harnesses & Reference Implementations > Coding-Agent Harnesses",Revo,https://github.com/shyringo/revo,,,,,,repository,"MIT terminal harness whose unit of continuity is one long-lived agent life rather than a task-scoped session: cognition runs without user input, sleep/wake persists unfinished work and deliveries, and user input preempts in-flight parallel work." walkinglabs_harness,279,"Runtimes, Harnesses & Reference Implementations > Coding-Agent Harnesses",UNCUT,https://github.com/Balanov/uncut,,,,,,repository,"Inspectable MIT-licensed Go coding-agent harness with full-history requests, untruncated tool results, resumable JSONL sessions, and a reproducible lossless proof suite." walkinglabs_harness,280,"Runtimes, Harnesses & Reference Implementations > Coding-Agent Harnesses",Synaps,https://github.com/HaseebKhalid1507/SynapsCLI,,,,,,repository,"Rust agent harness for terminal coding agents: parallel subagents, a session daemon that terminals can share and resume, MCP tools, and plugins, across any model." walkinglabs_harness,284,"Runtimes, Harnesses & Reference Implementations > Multi-Agent Orchestration",How we built our multi-agent research system,https://www.anthropic.com/engineering/multi-agent-research-system,,,,,,grey,Anthropic's architecture write-up for a multi-agent system with separation of roles and structured coordination. walkinglabs_harness,285,"Runtimes, Harnesses & Reference Implementations > Multi-Agent Orchestration",Concord MCP,https://github.com/Get-Concord-AI/concord-mcp,,,,,,repository,"MIT-licensed coordination protocol for coding agents across Claude Code, Codex, Cursor, Gemini CLI, and Grok Build, with explicit work claims, edit-overlap detection, durable messages, ownership transfer, and evidence-bearing handoffs exposed through five MCP workflow tools." walkinglabs_harness,286,"Runtimes, Harnesses & Reference Implementations > Multi-Agent Orchestration",Orkas,https://github.com/Orkas-AI/Orkas,,,,,,repository,"Local-first desktop harness for coordinating multiple agents with shared files, independent work contexts, human approval gates, and resumable execution." walkinglabs_harness,287,"Runtimes, Harnesses & Reference Implementations > Multi-Agent Orchestration",Agentlas OS,https://github.com/agentlas-ai/Agentlas-OS,,,,,,repository,"Local-first agent operation environment that composes specialist teams while retaining host-local tools, permissions, memory boundaries, and verification rules." walkinglabs_harness,288,"Runtimes, Harnesses & Reference Implementations > Multi-Agent Orchestration",Squadron,https://github.com/mlund01/squadron,,,,,,repository,"MIT-licensed declarative runtime for multi-agent workflows defined in HCL, including orchestration, state, dependency resolution, routing, persistence, and resume." walkinglabs_harness,289,"Runtimes, Harnesses & Reference Implementations > Multi-Agent Orchestration",Cowork Forge,https://github.com/sopaco/cowork-forge,,,,,,repository,MIT-licensed multi-agent software-development workflow with specialized roles and a staged pipeline from requirements through delivery. walkinglabs_harness,290,"Runtimes, Harnesses & Reference Implementations > Multi-Agent Orchestration",5dive,https://github.com/5dive-ai/5dive,,,,,,repository,"MIT multi-agent harness that uses the OS as its control layer: each coding-agent CLI (Claude Code, Codex, Hermes and others) runs as its own Linux user and systemd service, and agents hand off work through one bash CLI and a shared SQLite task queue." walkinglabs_harness,291,"Runtimes, Harnesses & Reference Implementations > Multi-Agent Orchestration",hcom,https://github.com/aannoo/hcom,,,,,,repository,"MIT-licensed coordination layer for coding-agent harnesses, with hook-driven messaging, activity observation, agent spawning, and event subscriptions across terminals." walkinglabs_harness,292,"Runtimes, Harnesses & Reference Implementations > Multi-Agent Orchestration",Raven,https://github.com/EverMind-AI/Raven,,,,,,repository,"Apache-2.0 host agent that plans complex tasks as DAGs and orchestrates built-in and third-party agents such as Claude Code and Codex over ACP, CLI, or OpenAI-compatible APIs; its agent loop is split into Memory, Planning, Capability, and Action strategy modules that an experimental Curator rewrites per agent, installing a change only after it is verified." walkinglabs_harness,296,"Runtimes, Harnesses & Reference Implementations > Browser, MCP & Tool Integration",browser-use/browser-harness,https://github.com/browser-use/browser-harness,,,,,,repository,"A thin CDP-based browser harness that lets agents extend helper functions during execution, useful for inspecting self-healing web-task workflows." walkinglabs_harness,297,"Runtimes, Harnesses & Reference Implementations > Browser, MCP & Tool Integration",Jet Browser,https://github.com/masakaai/jet-browser,,,,,,repository,"Apache-2.0 WPE WebKit runtime that gives an agent harness one isolated browser session per container, an ordered JSONL execution boundary, native input, semantic and visual evidence, and a reproducible offline acceptance check." walkinglabs_harness,298,"Runtimes, Harnesses & Reference Implementations > Browser, MCP & Tool Integration",Uni-CLI,https://github.com/olo-dot-io/Uni-CLI,,,,,,repository,"Universal CLI hub connecting agents to 134 sites and desktop apps via 711 declarative YAML pipelines. Ships an 8-phase Karpathy-style self-repair loop, eval harness with a starter catalog, per-call cost ledger, hardcoded sensitive-path deny list, and unicli mcp serve that auto-registers one MCP tool per adapter. ~80 tokens per invocation." walkinglabs_harness,299,"Runtimes, Harnesses & Reference Implementations > Browser, MCP & Tool Integration",OpenAgentRelay,https://github.com/ShakespeareLabs/open-agent-relay,,,,,,repository,"Inspectable runtime boundary for exposing a local agent or automation as a keyed LAN capability with target verification, bounded conversations, JSON output, and explicit exit codes." walkinglabs_harness,300,"Runtimes, Harnesses & Reference Implementations > Browser, MCP & Tool Integration",BrowserAct,https://github.com/browser-act/skills,,,,,,repository,"Open-source browser automation layer for agents with isolated parallel sessions, multi-account operation, and human handoff when automation is blocked." walkinglabs_harness,301,"Runtimes, Harnesses & Reference Implementations > Browser, MCP & Tool Integration",Cortex,https://github.com/cortex-docs/cortex,,,,,,repository,"Generates project-specific MCP servers from API contracts and Markdown, with typed tools, specifications, SDK guidance, and project documentation kept together as inspectable source." walkinglabs_harness,305,"Runtimes, Harnesses & Reference Implementations > Workflow, Profiles & Asset Management",Bring Your AI MCP,https://github.com/unitedideas/bringyour-mcp,,,,,,repository,"Public harness-migration reference for Claude Code to Codex moves, with installable auditor artifacts and explicit validation notes for hooks, MCP config, and instruction-file differences." walkinglabs_harness,306,"Runtimes, Harnesses & Reference Implementations > Workflow, Profiles & Asset Management",skills.sh,https://skills.sh,,,,,,grey,"A community marketplace for discovering, sharing, and installing reusable AI agent skills across runtimes like Claude Code and OpenClaw, making harness capabilities portable and composable." walkinglabs_harness,307,"Runtimes, Harnesses & Reference Implementations > Workflow, Profiles & Asset Management",stelow,https://github.com/calionauta/stelow,,,,,,repository,"Agentic product-workflow harness with Shape Up boundaries, adversarial plan review, acceptance-based execution contracts, and audit loops." walkinglabs_harness,308,"Runtimes, Harnesses & Reference Implementations > Workflow, Profiles & Asset Management",Build A Harness,https://github.com/3IVIS/buildaharness,,,,,,repository,Apache-2.0 visual canvas for agent harnesses that compiles a runtime-neutral FlowSpec to several orchestration frameworks. walkinglabs_harness,309,"Runtimes, Harnesses & Reference Implementations > Workflow, Profiles & Asset Management",agent-harness,https://github.com/ar27111994/agent-harness,,,,,,repository,"Reproducible lifecycle for coding-agent assets with authority-ranked discovery, pinned mirrors, quarantine routing, staged activation, and host-specific wiring." walkinglabs_harness,310,"Runtimes, Harnesses & Reference Implementations > Workflow, Profiles & Asset Management",codex-profiles,https://github.com/Ducksss/codex-profiles,,,,,,repository,"Codex CLI and Desktop profile launcher that isolates authentication, configuration, sessions, connectors, plugins, and logs by CODEX_HOME." walkinglabs_harness,311,"Runtimes, Harnesses & Reference Implementations > Workflow, Profiles & Asset Management",AgentPlane,https://github.com/basilisk-labs/agentplane,,,,,,repository,"Git-native workflow-control harness that stores task records, policy, verification evidence, and closure state as reviewable repository artifacts." walkinglabs_harness,312,"Runtimes, Harnesses & Reference Implementations > Workflow, Profiles & Asset Management",SAKO,https://github.com/Thakay/sako,,,,,,repository,"Repo-local task ledger and finish gate for Claude Code and Codex sessions: tasks, ownership and check evidence live in plain Markdown files, and SessionStart and Stop hooks check each session in and flag unrecorded work, stale or missing check results, and uncommitted changes before it ends." walkinglabs_harness,313,"Runtimes, Harnesses & Reference Implementations > Workflow, Profiles & Asset Management",harness-scope,https://github.com/shimo4228/harness-scope,,,,,,repository,"MIT Claude Code mod that scopes a global harness per repo: named profiles in ~/.claude decide which skills, agents, instruction files and tools each repo loads, with no network requests or model calls." picrew_harness,15,Featured Harness Blogs,Scaling Managed Agents: Decoupling the brain from the hands,https://www.anthropic.com/engineering/managed-agents,,,,,,grey,"Anthropic's meta-harness architecture for decoupling session logs, harness loops, and sandboxes in long-horizon agents." picrew_harness,16,Featured Harness Blogs,What We Learned Building Cloud Agents,https://cognition.ai/blog/what-we-learned-building-cloud-agents,,,,,,grey,"Cognition's field report on secure cloud-agent infrastructure, VM isolation, full-state snapshots, orchestration, governance, integrations, and enterprise adoption." picrew_harness,17,Featured Harness Blogs,Claude Code auto mode,https://www.anthropic.com/engineering/claude-code-auto-mode,,,,,,grey,Anthropic's write-up on classifier-backed approval delegation for safer high-autonomy coding-agent runs. picrew_harness,18,Featured Harness Blogs,Harness engineering (OpenAI),https://openai.com/index/harness-engineering/,,,,,,grey,](https://openai.com/index/harness-engineering/): Field report on building reliable agent-first software via harness constraints and verification. picrew_harness,19,Featured Harness Blogs,The next evolution of the Agents SDK,https://openai.com/index/the-next-evolution-of-the-agents-sdk/,,,,,,grey,"OpenAI's product and engineering post on model-native agent harnesses, native sandbox execution, manifests, memory, and filesystem and shell tools." picrew_harness,20,Featured Harness Blogs,Building Effective AI Agents,https://www.anthropic.com/engineering/building-effective-agents,,,,,,grey,Anthropic's practical guidance on when to use workflows vs. autonomous agents and how to structure them. picrew_harness,21,Featured Harness Blogs,Writing effective tools for AI agents,https://www.anthropic.com/engineering/writing-tools-for-agents,,,,,,grey,Best practices for tool interface design so agents call tools safely and reliably. picrew_harness,22,Featured Harness Blogs,Effective harnesses for long-running agents,https://www.anthropic.com/engineering/effective-harnesses-for-long-running-agents,,,,,,grey,"Practical guide to maintaining state, resumability, and reliability over long agent runs." picrew_harness,23,Featured Harness Blogs,Harness design for long-running application development,https://www.anthropic.com/engineering/harness-design-long-running-apps,,,,,,grey,Follow-up article on improving long-running app generation through harness structure. picrew_harness,24,Featured Harness Blogs,Improving Deep Agents with harness engineering,https://blog.langchain.com/improving-deep-agents-with-harness-engineering/,,,,,,grey,Evidence that harness improvements alone can move benchmark performance. picrew_harness,25,Featured Harness Blogs,Evaluating Deep Agents: Our Learnings,https://blog.langchain.com/evaluating-deep-agents-our-learnings/,,,,,,grey,LangChain's practical lessons on evaluating stateful and long-horizon agents. picrew_harness,26,Featured Harness Blogs,"Your Agent Needs a Harness, Not a Framework",https://www.inngest.com/blog/your-agent-needs-a-harness-not-a-framework,,,,,,grey,Argument for reliability-first infrastructure around agents instead of framework-only thinking. picrew_harness,27,Featured Harness Blogs,How GPT-5.6 fuses frontier intelligence with frontier efficiency,https://openai.com/index/gpt-5-6-frontier-intelligence-efficiency/,,,,,,grey,"OpenAI's engineering account of the Rust agentic harness behind Codex and ChatGPT Work, covering deferred tool discovery, bounded outputs, deterministic tool ordering, and prompt-cache preservation." picrew_harness,28,Featured Harness Blogs,How we contain Claude across products,https://www.anthropic.com/engineering/how-we-contain-claude,,,,,,grey,"Anthropic's cross-product containment architecture for claude.ai, Claude Code, and Cowork, spanning sandboxes, VMs, filesystem boundaries, egress controls, and approval fatigue." picrew_harness,29,Featured Harness Blogs,"Introducing Agent Executor, Google’s distributed Agent Runtime",https://cloud.google.com/blog/products/ai-machine-learning/agent-executor-googles-distributed-agent-runtime/,,,,,,grey,"Google's distributed agent runtime design for durable execution, event-log snapshots, secure isolation, session consistency, reconnection, and trajectory branching." picrew_harness,30,Featured Harness Blogs,"Building a safe, effective sandbox to enable Codex on Windows",https://openai.com/index/building-codex-windows-sandbox/,,,,,,grey,"OpenAI's implementation report on enforcing filesystem and network boundaries for Codex on Windows with restricted tokens, SIDs, and a dedicated command runner." picrew_harness,31,Featured Harness Blogs,Running Codex safely at OpenAI,https://openai.com/index/running-codex-safely/,,,,,,grey,"OpenAI's operational blueprint for governing Codex with sandbox and approval policies, network controls, managed identity, configuration, telemetry, and audit trails." picrew_harness,76,Catalog > Harness Architecture & Orchestration,Superpowers,https://github.com/obra/superpowers,,,,,,repository,"Cross-agent software development methodology built from composable skills, mandatory workflows, worktrees, planning, TDD, review, and subagent execution." picrew_harness,77,Catalog > Harness Architecture & Orchestration,Matt Pocock Skills,https://github.com/mattpocock/skills,,,,,,repository,"Composable engineering skills for Claude Code, Codex, and other coding agents, covering alignment, shared language, ADRs, specs, tests, and code review loops." picrew_harness,79,Catalog > Harness Architecture & Orchestration,Langflow,https://github.com/langflow-ai/langflow,,,,,,repository,"Visual agent workflow platform with multi-agent orchestration, conversation management, stepwise testing, API and MCP deployment, and tracing integrations." picrew_harness,80,Catalog > Harness Architecture & Orchestration,gstack,https://github.com/garrytan/gstack,,,,,,repository,"Claude Code and cross-agent skill stack that turns product planning, architecture review, QA, security, release, and retrospectives into repeatable agent workflows." picrew_harness,81,Catalog > Harness Architecture & Orchestration,Addy's Agent Skills,https://github.com/addyosmani/agent-skills,,,,,,repository,"Production-grade engineering skills for coding agents that package lifecycle workflows, quality gates, reviews, testing, debugging, security, and release practices." picrew_harness,82,Catalog > Harness Architecture & Orchestration,DeerFlow,https://github.com/bytedance/deer-flow,,,,,,repository,"Long-horizon super-agent harness integrating memory, tools, subagents, and sandboxes." picrew_harness,83,Catalog > Harness Architecture & Orchestration,Orca,https://github.com/stablyai/orca,,,,,,repository,"Cross-agent desktop and mobile control plane for running CLI coding agents in isolated worktrees, monitoring their state, reviewing changes, and steering remote or local sessions." picrew_harness,84,Catalog > Harness Architecture & Orchestration,Ruflo,https://github.com/ruvnet/ruflo,,,,,,repository,"Multi-agent orchestration platform for Claude Code with swarms, persistent memory, federation, plugins, and MCP hooks." picrew_harness,85,Catalog > Harness Architecture & Orchestration,oh-my-openagent,https://github.com/code-yeongyu/oh-my-openagent,,,,,,repository,"Multi-harness agent OS for OpenCode, Codex, Claude Code, and other coding agents with team-mode orchestration, background agents, MCPs, and skills." picrew_harness,86,Catalog > Harness Architecture & Orchestration,AutoGen,https://github.com/microsoft/autogen,,,,,,repository,Programming framework for agentic AI with multi-agent interaction and orchestration. picrew_harness,87,Catalog > Harness Architecture & Orchestration,CrewAI,https://github.com/crewAIInc/crewAI,,,,,,repository,"Multi-agent automation framework with production Flows, autonomous Crews, event-driven control, tracing, guardrails, memory, and human review hooks." picrew_harness,88,Catalog > Harness Architecture & Orchestration,Agno,https://github.com/agno-agi/agno,,,,,,repository,Agent software runtime focused on running and managing agentic systems at scale. picrew_harness,89,Catalog > Harness Architecture & Orchestration,LangGraph,https://github.com/langchain-ai/langgraph,,,,,,repository,Graph-based runtime for resilient stateful agents and deterministic workflow control. picrew_harness,90,Catalog > Harness Architecture & Orchestration,Herdr,https://github.com/herdrdev/herdr,,,,,,repository,"Persistent terminal runtime for coding agents with reconnectable sessions, blocked and idle detection, agent-driven orchestration APIs, and plugin-extensible workflows." picrew_harness,91,Catalog > Harness Architecture & Orchestration,Claude Code Router,https://github.com/musistudio/claude-code-router,,,,,,repository,"Local model gateway and control plane for coding agents with provider profiles, conditional routing, credential pools, retries, fallbacks, tool extensions, quotas, and request traces." picrew_harness,92,Catalog > Harness Architecture & Orchestration,ChatDev,https://github.com/OpenBMB/ChatDev,,,,,,repository,"Zero-code multi-agent orchestration platform for defining agents, workflows, and tasks, evolved from a virtual software-company harness." picrew_harness,93,Catalog > Harness Architecture & Orchestration,Buzz,https://github.com/block/buzz,,,,,,repository,"Self-hostable human-agent workspace built on a signed event log, with scoped identities, audit trails, agent-first CLI and ACP bridges, workflows, Git events, search, and approvals." picrew_harness,94,Catalog > Harness Architecture & Orchestration,AionUi,https://github.com/iOfficeAI/AionUi,,,,,,repository,"Cross-platform cowork control plane with a built-in agent engine, external CLI-agent orchestration, team mode, per-agent approvals, remote channels, MCP management, and scheduled runs." picrew_harness,95,Catalog > Harness Architecture & Orchestration,deepagents,https://github.com/langchain-ai/deepagents,,,,,,repository,"Open-source harness for long-running, tool-using agents with planning and subagent patterns." picrew_harness,96,Catalog > Harness Architecture & Orchestration,OpenAI Agents SDK (Python),https://github.com/openai/openai-agents-python,,,,,,repository,"Lightweight framework for multi-agent workflows, handoffs, and production patterns." picrew_harness,97,Catalog > Harness Architecture & Orchestration,Semantic Kernel,https://github.com/microsoft/semantic-kernel,,,,,,repository,Enterprise-grade agentic application framework with orchestration and plugin patterns. picrew_harness,98,Catalog > Harness Architecture & Orchestration,Symphony,https://github.com/openai/symphony,,,,,,repository,Ticket-driven orchestration layer that turns project work into isolated autonomous implementation runs. picrew_harness,99,Catalog > Harness Architecture & Orchestration,Archon,https://github.com/coleam00/Archon,,,,,,repository,"Workflow engine for AI coding agents with YAML-defined phases, isolated worktrees, and validation gates." picrew_harness,100,Catalog > Harness Architecture & Orchestration,Go Micro,https://github.com/micro/go-micro,,,,,,repository,"Go agent harness and service framework combining model loops, durable memory, tools, planning, delegation, guardrails, service discovery, MCP, A2A, and durable flows." picrew_harness,101,Catalog > Harness Architecture & Orchestration,Google ADK (Python),https://github.com/google/adk-python,,,,,,repository,"Code-first toolkit to build, evaluate, and deploy advanced AI agents." picrew_harness,102,Catalog > Harness Architecture & Orchestration,PydanticAI,https://github.com/pydantic/pydantic-ai,,,,,,repository,Type-safe Python framework for agents with strong schema contracts and tooling. picrew_harness,103,Catalog > Harness Architecture & Orchestration,elizaOS,https://github.com/elizaOS/eliza,,,,,,repository,"Extensible agent runtime and operating system with CLI scaffolding, agent loop, plugins, memory/state primitives, dashboards, connectors, and benchmark suites." picrew_harness,104,Catalog > Harness Architecture & Orchestration,Gas Town,https://github.com/gastownhall/gastown,,,,,,repository,"Multi-agent workspace manager for coordinating coding agents with persistent work tracking, git-backed hooks, handoffs, supervision, and merge queues." picrew_harness,105,Catalog > Harness Architecture & Orchestration,CAMEL,https://github.com/camel-ai/camel,,,,,,repository,"Multi-agent framework for large-scale agent systems with communication, stateful memory, environments, benchmarks, and coordination primitives." picrew_harness,107,Catalog > Harness Architecture & Orchestration,CloudCLI,https://github.com/siteboon/claudecodeui,,,,,,repository,"Web and mobile control surface for Claude Code, OpenCode, Cursor CLI, Codex, and Gemini CLI with project, terminal, file, Git, session, and plugin management." picrew_harness,108,Catalog > Harness Architecture & Orchestration,Microsoft Agent Framework,https://github.com/microsoft/agent-framework,,,,,,repository,"Multi-language framework for building, orchestrating, and deploying AI agents with graph workflows and observability." picrew_harness,109,Catalog > Harness Architecture & Orchestration,Hive,https://github.com/aden-hive/hive,,,,,,repository,Outcome-driven agent runtime harness with explicit control loops and orchestration blocks. picrew_harness,110,Catalog > Harness Architecture & Orchestration,VoltAgent,https://github.com/VoltAgent/voltagent,,,,,,repository,TypeScript agent engineering platform built around open runtime abstractions. picrew_harness,111,Catalog > Harness Architecture & Orchestration,Omnigent,https://github.com/omnigent-ai/omnigent,,,,,,repository,"Open-source meta-harness that orchestrates Claude Code, Codex, Cursor, Pi, and custom agents with shared sessions, cloud sandboxes, policies, and collaboration." picrew_harness,112,Catalog > Harness Architecture & Orchestration,PraisonAI,https://github.com/MervinPraison/PraisonAI,,,,,,repository,"Multi-agent workforce framework with autonomous planning, execution, memory, RAG, dashboards, and multi-provider model support." picrew_harness,113,Catalog > Harness Architecture & Orchestration,mcp-agent,https://github.com/lastmile-ai/mcp-agent,,,,,,repository,Practical agent framework centered on MCP tool ecosystems and workflow composition. picrew_harness,114,Catalog > Harness Architecture & Orchestration,Claude Squad,https://github.com/smtg-ai/claude-squad,,,,,,repository,"Terminal control plane for managing multiple Claude Code, Codex, Gemini, OpenCode, Amp, and other agent sessions with isolated git workspaces and review flows." picrew_harness,115,Catalog > Harness Architecture & Orchestration,Flue,https://github.com/withastro/flue,,,,,,repository,"TypeScript harness framework for building headless agents with sessions, tools, skills, and pluggable sandboxes." picrew_harness,117,Catalog > Harness Architecture & Orchestration,Agent Squad,https://github.com/2FastLabs/agent-squad,,,,,,repository,"Multi-agent orchestration framework that routes requests, preserves conversation context, supports Python/TypeScript, and coordinates specialist agents." picrew_harness,118,Catalog > Harness Architecture & Orchestration,Strands Agents,https://github.com/strands-agents/harness-sdk,,,,,,repository,"Model-driven agent SDK and monorepo with Python/TypeScript agent loops, provider adapters, tools, MCP integration, multi-agent systems, and streaming." picrew_harness,119,Catalog > Harness Architecture & Orchestration,Open Multi-Agent,https://github.com/open-multi-agent/open-multi-agent,,,,,,repository,"TypeScript-native multi-agent orchestrator that turns goals into task DAGs with parallel execution, MCP integration, and live tracing." picrew_harness,120,Catalog > Harness Architecture & Orchestration,AIOS,https://github.com/agiresearch/AIOS,,,,,,repository,"AI Agent Operating System with a kernel and SDK for scheduling, context management, memory, storage, tools, deployment modes, and computer-use sandboxing." picrew_harness,121,Catalog > Harness Architecture & Orchestration,Nexent,https://github.com/ModelEngine-Group/nexent,,,,,,repository,"Zero-code agent platform built around harness engineering principles, unifying tools, skills, memory, orchestration, constraints, feedback loops, and control planes." picrew_harness,122,Catalog > Harness Architecture & Orchestration,Cloudflare Agents,https://github.com/cloudflare/agents,,,,,,repository,Platform runtime for building and deploying agents with production infrastructure primitives. picrew_harness,123,Catalog > Harness Architecture & Orchestration,Embabel Agent Framework,https://github.com/embabel/embabel-agent,,,,,,repository,"JVM agent framework for typed agentic flows with goals, actions, conditions, dynamic planning, platform modes, and testability." picrew_harness,124,Catalog > Harness Architecture & Orchestration,cascadeflow,https://github.com/lemony-ai/cascadeflow,,,,,,repository,"In-process agent harness for per-step model routing, tool budget gates, runtime policy actions, and auditable control inside agent loops." picrew_harness,125,Catalog > Harness Architecture & Orchestration,OpenAI Agents SDK (JS/TS),https://github.com/openai/openai-agents-js,,,,,,repository,"JavaScript/TypeScript framework for multi-agent workflows with handoffs, tools, guardrails, sessions, tracing, and sandbox agents." picrew_harness,126,Catalog > Harness Architecture & Orchestration,Reef,https://github.com/Human-Agent-Society/reef,,,,,,repository,"Continual learning infrastructure connecting agent inference, feedback, evaluation, and versioned delivery to improve model weights, prompts, rules, and skills." picrew_harness,127,Catalog > Harness Architecture & Orchestration,Docker Agent,https://github.com/docker/docker-agent,,,,,,repository,Agent builder and runtime stack emphasizing container-native execution. picrew_harness,128,Catalog > Harness Architecture & Orchestration,NeMo Agent Toolkit,https://github.com/NVIDIA/NeMo-Agent-Toolkit,,,,,,repository,Open toolkit for connecting and optimizing teams of AI agents. picrew_harness,129,Catalog > Harness Architecture & Orchestration,Apache Burr,https://github.com/apache/burr,,,,,,repository,"State-machine framework for decision-making agents and LLM apps with persistence, telemetry UI, tracing, and framework-agnostic execution." picrew_harness,131,Catalog > Harness Architecture & Orchestration,Shannon,https://github.com/Kocoro-lab/Shannon,,,,,,repository,"Production-oriented multi-agent orchestration framework with Temporal workflows, WASI sandboxing, OPA approvals, token budgets, and OpenTelemetry tracing." picrew_harness,132,Catalog > Harness Architecture & Orchestration,tRPC-Agent-Go,https://github.com/trpc-group/trpc-agent-go,,,,,,repository,"Go framework for production agent systems with graph workflows, tools, memory, evaluation, protocols, and OpenTelemetry observability." picrew_harness,133,Catalog > Harness Architecture & Orchestration,Scion,https://github.com/GoogleCloudPlatform/scion,,,,,,repository,"Experimental multi-agent orchestration testbed that runs isolated agent harnesses in containers, worktrees, and remote runtimes." picrew_harness,134,Catalog > Harness Architecture & Orchestration,deepagentsjs,https://github.com/langchain-ai/deepagentsjs,,,,,,repository,"TypeScript agent harness with built-in planning, filesystem tools, subagents, and LangGraph-native runtime hooks." picrew_harness,135,Catalog > Harness Architecture & Orchestration,oh-my-agent,https://github.com/first-fluke/oh-my-agent,,,,,,repository,"Portable multi-agent harness that projects shared agents, skills, workflows, and rules into multiple coding-agent runtimes." picrew_harness,136,Catalog > Harness Architecture & Orchestration,LiteLLM Agent Control Plane,https://github.com/LiteLLM-Labs/litellm-agent-control-plane,,,,,,repository,"Agent control plane that provides one API and UI across OpenCode, Hermes, Claude Managed Agents, Cursor Agents API, DeepAgents, and OpenClaw runtimes." picrew_harness,137,Catalog > Harness Architecture & Orchestration,Chorus,https://github.com/Chorus-AIDLC/Chorus,,,,,,repository,"AI-human collaboration harness for session lifecycle, task state, sub-agent orchestration, observability, and recovery." picrew_harness,138,Catalog > Harness Architecture & Orchestration,Pydantic AI Harness,https://github.com/pydantic/pydantic-ai-harness,,,,,,repository,"Official Pydantic AI capability library for composing tools, lifecycle hooks, instructions, and model settings into reusable agent harnesses." picrew_harness,139,Catalog > Harness Architecture & Orchestration,Water,https://github.com/manthanguptaa/water,,,,,,repository,"Python agent harness framework for orchestration, resilience, observability, guardrails, approval gates, sandboxing, and deployment." picrew_harness,140,Catalog > Harness Architecture & Orchestration,OmniCoreAgent,https://github.com/omnirexflora-labs/omnicoreagent,,,,,,repository,"Python production harness with model loop, tools, MCP, memory, workspace files, guardrails, events, subagents, background tasks, and REST/SSE serving." picrew_harness,141,Catalog > Harness Architecture & Orchestration,hankweave,https://github.com/SouthBridgeAI/hankweave-runtime,,,,,,repository,"Headless-first long-horizon runtime that orchestrates existing agent harnesses with sentinels, loops, checkpoints, and event journals." picrew_harness,148,Catalog > Context & Working-State Engineering,Graphify,https://github.com/Graphify-Labs/graphify,,,,,,repository,"Cross-agent skill and CLI that turns code, docs, schemas, and media into a queryable knowledge graph with Codex and Claude Code hooks." picrew_harness,149,Catalog > Context & Working-State Engineering,claude-mem,https://github.com/thedotmack/claude-mem,,,,,,repository,Plugin-style memory layer that captures session history and reinjects relevant context into future coding runs. picrew_harness,150,Catalog > Context & Working-State Engineering,Headroom,https://github.com/headroomlabs-ai/headroom,,,,,,repository,"Context-compression layer for agent tool outputs, logs, files, RAG chunks, and conversations, available as a library, proxy, agent wrapper, and MCP server." picrew_harness,151,Catalog > Context & Working-State Engineering,CodeGraph,https://github.com/colbymchenry/codegraph,,,,,,repository,"Local code knowledge-graph MCP server and installer that wires semantic code intelligence into Claude Code, Codex, Gemini, Cursor, OpenCode, and other agents." picrew_harness,152,Catalog > Context & Working-State Engineering,Mem0,https://github.com/mem0ai/mem0,,,,,,repository,"Universal memory layer for AI agents with user, session, and agent state, hybrid retrieval, SDKs, and self-hosted deployment." picrew_harness,153,Catalog > Context & Working-State Engineering,MemPalace,https://github.com/MemPalace/mempalace,,,,,,repository,"Local-first verbatim memory layer with scoped semantic retrieval, pluggable storage backends, conversation mining, MCP integration, and reproducible LongMemEval results." picrew_harness,154,Catalog > Context & Working-State Engineering,OpenViking,https://github.com/volcengine/OpenViking,,,,,,repository,"Context database for agents that unifies memory, resources, and skills in a virtual filesystem with tiered loading, recursive retrieval, observable trajectories, and session-to-memory extraction." picrew_harness,155,Catalog > Context & Working-State Engineering,code-review-graph,https://github.com/tirth8205/code-review-graph,,,,,,repository,"Local-first code intelligence graph that incrementally maps dependencies and exposes targeted review context, blast-radius analysis, and risk-scored CI gates through MCP and CLI." picrew_harness,156,Catalog > Context & Working-State Engineering,Cognee,https://github.com/topoteretes/cognee,,,,,,repository,"Persistent agent memory platform combining knowledge graphs, vector retrieval, session distillation, feedback, and MCP or SDK integration across runs." picrew_harness,157,Catalog > Context & Working-State Engineering,agentmemory,https://github.com/rohitg00/agentmemory,,,,,,repository,"Persistent memory server for coding agents using hooks, MCP/REST integration, hybrid search, and shared session recall." picrew_harness,158,Catalog > Context & Working-State Engineering,Beads,https://github.com/gastownhall/beads,,,,,,repository,Agent-optimized distributed issue tracker that stores long-horizon coding work as dependency-aware graph state with memory recall and multi-branch sync. picrew_harness,159,Catalog > Context & Working-State Engineering,TencentDB Agent Memory,https://github.com/TencentCloud/TencentDB-Agent-Memory,,,,,,repository,"Local agent memory plugin combining symbolic short-term state, layered long-term memory, traceability, and OpenClaw/Hermes integrations." picrew_harness,160,Catalog > Context & Working-State Engineering,planning-with-files,https://github.com/OthmanAdi/planning-with-files,,,,,,repository,Skill package for persistent file-based planning in coding-agent workflows. picrew_harness,161,Catalog > Context & Working-State Engineering,Hindsight,https://github.com/vectorize-io/hindsight,,,,,,repository,"Production agent memory system with retain, recall, and reflect operations, isolated memory banks, evidence-backed observations, coding-agent integrations, MCP, and memory-defense policies." picrew_harness,162,Catalog > Context & Working-State Engineering,Context Mode,https://github.com/mksglu/context-mode,,,,,,repository,"MCP context optimization server that sandboxes tool output, indexes session events, and restores continuity across agent compactions." picrew_harness,163,Catalog > Context & Working-State Engineering,Agent Skills for Context Engineering,https://github.com/muratcankoylan/Agent-Skills-for-Context-Engineering,,,,,,repository,Large skill library oriented around context engineering and production agents. picrew_harness,164,Catalog > Context & Working-State Engineering,SkillOpt,https://github.com/microsoft/SkillOpt,,,,,,repository,"Microsoft optimizer that trains reusable natural-language agent skills through trajectory edits, validation gates, and deployable skill artifacts." picrew_harness,165,Catalog > Context & Working-State Engineering,Memori,https://github.com/MemoriLabs/Memori,,,,,,repository,"Agent memory layer with entity and process attribution, session capture, persistent recall, context injection, and SDK, MCP, OpenClaw, and Hermes integrations." picrew_harness,166,Catalog > Context & Working-State Engineering,Trellis,https://github.com/mindfold-ai/Trellis,,,,,,repository,"Multi-platform coding-agent workflow framework with task context, project memory, and spec injection." picrew_harness,167,Catalog > Context & Working-State Engineering,memU,https://github.com/NevaMind-AI/memU,,,,,,repository,"Memory harness for AI agents that turns raw workspace data into structured, queryable context for agent retrieval." picrew_harness,168,Catalog > Context & Working-State Engineering,EverOS,https://github.com/EverMind-AI/EverOS,,,,,,repository,"Portable self-evolving memory runtime for agents, persisting readable Markdown with local SQLite/LanceDB indexes and reusable cases and skills." picrew_harness,169,Catalog > Context & Working-State Engineering,MemOS,https://github.com/MemTensor/MemOS,,,,,,repository,"Agent memory infrastructure with editable memory graphs, isolated or shared memory cubes, asynchronous ingestion, and local hybrid-retrieval plugins for OpenClaw, Hermes, and DeepSeek Harness." picrew_harness,170,Catalog > Context & Working-State Engineering,Context-Engineering Handbook,https://github.com/davidkimai/Context-Engineering,,,,,,repository,First-principles handbook focused on practical context engineering for agent systems. picrew_harness,171,Catalog > Context & Working-State Engineering,CCPM,https://github.com/automazeio/ccpm,,,,,,repository,Spec-driven project-manager skill that turns PRDs and GitHub issues into persistent context and parallel agent execution. picrew_harness,172,Catalog > Context & Working-State Engineering,Engram,https://github.com/Gentleman-Programming/engram,,,,,,repository,"Agent-agnostic persistent memory system packaged as a Go binary with SQLite/FTS5, MCP server, HTTP API, CLI, TUI, and sync workflows." picrew_harness,173,Catalog > Context & Working-State Engineering,ByteRover CLI,https://github.com/campfirein/byterover-cli,,,,,,repository,"Portable context-memory CLI for coding agents with a context tree, review workflow, versioning, built-in tools, cloud sync, and MCP integration." picrew_harness,174,Catalog > Context & Working-State Engineering,Acontext,https://github.com/memodb-io/Acontext,,,,,,repository,Skill-memory layer that distills agent runs into inspectable skill files and recalls them through agent-controlled tools. picrew_harness,175,Catalog > Context & Working-State Engineering,Awesome Context Engineering,https://github.com/Meirtz/Awesome-Context-Engineering,,,,,,repository,Survey-style list for context engineering resources and frameworks. picrew_harness,176,Catalog > Context & Working-State Engineering,agentic-stack,https://github.com/codejunkie99/agentic-stack,,,,,,repository,"Portable memory, skills, protocols, and dashboard layer that keeps state across multiple coding-agent harnesses." picrew_harness,177,Catalog > Context & Working-State Engineering,context-space,https://github.com/context-space/context-space,,,,,,repository,Infrastructure project focused on context engineering building blocks and MCP-centric integrations. picrew_harness,178,Catalog > Context & Working-State Engineering,Memorix,https://github.com/AVIDS2/memorix,,,,,,repository,"Local-first cross-agent memory control plane with MCP support, workspace sync, sessions, and orchestration state." picrew_harness,179,Catalog > Context & Working-State Engineering,sd0x-dev-flow,https://github.com/sd0xdev/sd0x-dev-flow,,,,,,repository,"Claude Code harness layer with hook-enforced dual review, durable state-machine gates, context-compaction recovery, and fail-closed safety." picrew_harness,186,Catalog > Execution Substrates & Sandboxing,Daytona,https://github.com/daytonaio/daytona,,,,,,repository,"Secure and elastic sandbox infrastructure for running AI-generated code with file, Git, LSP, and execution APIs." picrew_harness,187,Catalog > Execution Substrates & Sandboxing,agent-browser,https://github.com/vercel-labs/agent-browser,,,,,,repository,"Native browser automation CLI for AI agents with accessibility snapshots, semantic locators, screenshots, streaming, and batch execution." picrew_harness,189,Catalog > Execution Substrates & Sandboxing,Browser Harness,https://github.com/browser-use/browser-harness,,,,,,repository,Thin editable CDP harness that connects LLMs directly to real browsers and lets agents extend helpers in flight. picrew_harness,190,Catalog > Execution Substrates & Sandboxing,OpenSandbox,https://github.com/opensandbox-group/OpenSandbox,,,,,,repository,Secure and extensible sandbox runtime built for agent workloads. picrew_harness,192,Catalog > Execution Substrates & Sandboxing,InsForge,https://github.com/InsForge/InsForge,,,,,,repository,"Backend substrate for agentic coding that exposes database, auth, storage, compute, hosting, logs, and model-gateway operations through MCP and CLI/Skills." picrew_harness,193,Catalog > Execution Substrates & Sandboxing,CubeSandbox,https://github.com/TencentCloud/CubeSandbox,,,,,,repository,"MicroVM-based sandbox service for AI agents with sub-60ms startup, E2B-compatible APIs, and hardware-level isolation." picrew_harness,194,Catalog > Execution Substrates & Sandboxing,OpenShell,https://github.com/NVIDIA/OpenShell,,,,,,repository,"Safe private runtime for autonomous agents with sandbox lifecycle control and declarative filesystem, network, process, and inference policies." picrew_harness,195,Catalog > Execution Substrates & Sandboxing,Microsandbox,https://github.com/superradcompany/microsandbox,,,,,,repository,"Rootless local VM sandbox runtime with SDKs, detached long-running sessions, agent skills, and MCP server integration." picrew_harness,196,Catalog > Execution Substrates & Sandboxing,Sandcastle,https://github.com/mattpocock/sandcastle,,,,,,repository,TypeScript library for orchestrating coding agents inside isolated sandboxes with configurable branch strategies. picrew_harness,197,Catalog > Execution Substrates & Sandboxing,agent-infra sandbox,https://github.com/agent-infra/sandbox,,,,,,repository,"All-in-one sandbox combining browser, shell, files, MCP, and IDE server." picrew_harness,198,Catalog > Execution Substrates & Sandboxing,Judge0,https://github.com/judge0/judge0,,,,,,repository,Scalable sandboxed code execution system usable as an agent execution backend. picrew_harness,199,Catalog > Execution Substrates & Sandboxing,Agent Sandbox,https://github.com/kubernetes-sigs/agent-sandbox,,,,,,repository,"Kubernetes-native sandbox control plane for isolated, stateful agent runtimes with stable identity, persistence, and warm-pool support." picrew_harness,200,Catalog > Execution Substrates & Sandboxing,stakpak/agent,https://github.com/stakpak/agent,,,,,,repository,Always-on open agent that runs on your machines with autonomous operational loops. picrew_harness,201,Catalog > Execution Substrates & Sandboxing,Sandbox Agent,https://github.com/rivet-dev/sandbox-agent,,,,,,repository,"HTTP/SSE control server for running coding agents inside sandboxes with normalized sessions, permissions, event streaming, and replay." picrew_harness,202,Catalog > Execution Substrates & Sandboxing,E2B Desktop Sandbox,https://github.com/e2b-dev/desktop,,,,,,repository,Secure virtual desktop sandbox for computer-use agents with SDK control and screen streaming. picrew_harness,203,Catalog > Execution Substrates & Sandboxing,OSS-Fuzz Gen,https://github.com/google/oss-fuzz-gen,,,,,,repository,LLM-powered fuzzing workflows integrated with controlled execution contexts. picrew_harness,204,Catalog > Execution Substrates & Sandboxing,AgentBay SDK,https://github.com/agentbay-ai/wuying-agentbay-sdk,,,,,,repository,"Cloud sandbox SDK for agents spanning browser, desktop, mobile, and code execution environments." picrew_harness,205,Catalog > Execution Substrates & Sandboxing,Tensorlake,https://github.com/tensorlakeai/tensorlake,,,,,,repository,"Serverless runtime for agent sandboxes with MicroVM isolation, snapshots, suspend-resume, and background orchestration." picrew_harness,206,Catalog > Execution Substrates & Sandboxing,AgentScope Runtime,https://github.com/agentscope-ai/agentscope-runtime,,,,,,repository,"Production runtime for agent apps with secure tool sandboxes, deployment APIs, observability, and state services." picrew_harness,207,Catalog > Execution Substrates & Sandboxing,SWE-ReX,https://github.com/SWE-agent/SWE-ReX,,,,,,repository,Sandboxed execution infrastructure for AI coding agents at local and cloud scale. picrew_harness,208,Catalog > Execution Substrates & Sandboxing,sandboxed.sh,https://github.com/Th0rgal/sandboxed.sh,,,,,,repository,Self-hosted orchestrator running coding agents inside isolated Linux workspaces. picrew_harness,209,Catalog > Execution Substrates & Sandboxing,Capsule,https://github.com/capsulerun/capsule,,,,,,repository,Durable runtime that coordinates agent tasks inside isolated WebAssembly sandboxes with retries and lifecycle tracking. picrew_harness,210,Catalog > Execution Substrates & Sandboxing,agentbox,https://github.com/mattolson/agent-sandbox,,,,,,repository,"Locked-down local sandbox for AI coding agents with scoped filesystem access, egress policy, secret injection, firewalling, and persistent agent state." picrew_harness,211,Catalog > Execution Substrates & Sandboxing,HexAgent,https://github.com/UnicomAI/hexagent,,,,,,repository,"Agent harness that separates the runtime from the computer it operates on through local, VM, and cloud sandbox backends." picrew_harness,212,Catalog > Execution Substrates & Sandboxing,terminal-bench-env,https://github.com/ucsb-mlsec/terminal-bench-env,,,,,,repository,Environment layer for terminal-agent benchmark execution. picrew_harness,219,"Catalog > Protocols, Tool Interfaces & Agent Contracts",Anthropic Agent Skills,https://github.com/anthropics/skills,,,,,,repository,"Official Agent Skills repository containing the skills specification, templates, and reference skill implementations for Claude." picrew_harness,220,"Catalog > Protocols, Tool Interfaces & Agent Contracts",GitHub Spec Kit,https://github.com/github/spec-kit,,,,,,repository,Toolkit for spec-driven development to guide deterministic agent execution. picrew_harness,221,"Catalog > Protocols, Tool Interfaces & Agent Contracts",MCP Servers,https://github.com/modelcontextprotocol/servers,,,,,,repository,Official collection of MCP server implementations across tools and domains. picrew_harness,222,"Catalog > Protocols, Tool Interfaces & Agent Contracts",Chrome DevTools MCP,https://github.com/ChromeDevTools/chrome-devtools-mcp,,,,,,repository,"Official MCP server that gives coding agents Chrome DevTools access for reliable browser automation, debugging, and performance analysis." picrew_harness,223,"Catalog > Protocols, Tool Interfaces & Agent Contracts",AAS Core,https://github.com/sickn33/agentic-awesome-skills,,,,,,repository,"Local agent-first control plane for discovering, selecting, validating, and planning reproducible skill stacks through a read-only MCP interface, manifests, plugins, and host-specific installers." picrew_harness,224,"Catalog > Protocols, Tool Interfaces & Agent Contracts",Awesome Copilot,https://github.com/github/awesome-copilot,,,,,,repository,"GitHub-managed collection of Copilot agents, instructions, skills, hooks, workflows, plugins, tools, and machine-readable listings." picrew_harness,225,"Catalog > Protocols, Tool Interfaces & Agent Contracts",Playwright MCP,https://github.com/microsoft/playwright-mcp,,,,,,repository,Official Playwright MCP server giving agents structured accessibility snapshots and deterministic browser automation tools. picrew_harness,226,"Catalog > Protocols, Tool Interfaces & Agent Contracts",Claude Code Plugins Directory,https://github.com/anthropics/claude-plugins-official,,,,,,repository,"Anthropic-managed Claude Code plugin marketplace defining plugin manifests, MCP configuration, commands, agents, skills, and submission quality gates." picrew_harness,227,"Catalog > Protocols, Tool Interfaces & Agent Contracts",OpenAI Codex Plugin for Claude Code,https://github.com/openai/codex-plugin-cc,,,,,,repository,"Official Claude Code plugin that exposes Codex review, adversarial review, delegation, status, result, and cancel commands inside another coding-agent harness." picrew_harness,228,"Catalog > Protocols, Tool Interfaces & Agent Contracts",GitHub MCP Server,https://github.com/github/github-mcp-server,,,,,,repository,"GitHub's official MCP server connecting agents to repositories, issues, pull requests, CI workflows, releases, and code analysis tools." picrew_harness,229,"Catalog > Protocols, Tool Interfaces & Agent Contracts",skills,https://github.com/vercel-labs/skills,,,,,,repository,"CLI for installing, using, finding, updating, and initializing Agent Skills across OpenCode, Claude Code, Codex, Cursor, and other coding agents." picrew_harness,230,"Catalog > Protocols, Tool Interfaces & Agent Contracts",Vercel Agent Skills,https://github.com/vercel-labs/agent-skills,,,,,,repository,Vercel's official Agent Skills collection packaging reusable instructions and scripts in the open Agent Skills format for coding-agent workflows. picrew_harness,231,"Catalog > Protocols, Tool Interfaces & Agent Contracts",Serena,https://github.com/oraios/serena,,,,,,repository,"MCP toolkit that gives coding agents IDE-like semantic retrieval, editing, refactoring, debugging, and memory tools." picrew_harness,232,"Catalog > Protocols, Tool Interfaces & Agent Contracts",DESIGN.md,https://github.com/google-labs-code/design.md,,,,,,repository,"Format specification and CLI for giving coding agents persistent, machine-readable design tokens and human-readable design rationale." picrew_harness,233,"Catalog > Protocols, Tool Interfaces & Agent Contracts",FastMCP,https://github.com/PrefectHQ/fastmcp,,,,,,repository,"Python framework for building MCP servers and clients with generated schemas, validation, documentation, production deployment patterns, and governance hooks." picrew_harness,234,"Catalog > Protocols, Tool Interfaces & Agent Contracts",Agent Skills Specification,https://github.com/agentskills/agentskills,,,,,,repository,"Open specification and documentation for packaging reusable agent capabilities, workflows, scripts, references, and assets behind progressive disclosure." picrew_harness,235,"Catalog > Protocols, Tool Interfaces & Agent Contracts",AGENTS.md,https://github.com/agentsmd/agents.md,,,,,,repository,Open format for repository-local instructions that coding agents can follow. picrew_harness,236,"Catalog > Protocols, Tool Interfaces & Agent Contracts",MCP Python SDK,https://github.com/modelcontextprotocol/python-sdk,,,,,,repository,"Official Python implementation of MCP for building clients and servers that expose tools, resources, prompts, protocol lifecycle events, and standard transports." picrew_harness,237,"Catalog > Protocols, Tool Interfaces & Agent Contracts",Google Agent Skills,https://github.com/google/skills,,,,,,repository,"Official Google Agent Skills repository for Google products and technologies, including Agent Platform and Google Cloud workflows." picrew_harness,238,"Catalog > Protocols, Tool Interfaces & Agent Contracts",MCP Toolbox for Databases,https://github.com/googleapis/mcp-toolbox,,,,,,repository,"Google open-source MCP server and custom tools framework for connecting agents, IDEs, and applications to enterprise databases." picrew_harness,239,"Catalog > Protocols, Tool Interfaces & Agent Contracts",MCP TypeScript SDK,https://github.com/modelcontextprotocol/typescript-sdk,,,,,,repository,"Official TypeScript MCP SDK with server and client packages, transports, auth helpers, middleware adapters, and runnable examples." picrew_harness,240,"Catalog > Protocols, Tool Interfaces & Agent Contracts",Hugging Face Skills,https://github.com/huggingface/skills,,,,,,repository,"Official Hugging Face Agent Skills collection for packaging Hub, dataset, training, evaluation, Spaces, and tooling workflows across major coding agents." picrew_harness,241,"Catalog > Protocols, Tool Interfaces & Agent Contracts",MCP Inspector,https://github.com/modelcontextprotocol/inspector,,,,,,repository,Official developer tool for testing and debugging MCP servers through an interactive client UI and protocol bridge. picrew_harness,242,"Catalog > Protocols, Tool Interfaces & Agent Contracts",mcp-use,https://github.com/mcp-use/mcp-use,,,,,,repository,Full-stack MCP framework for building MCP apps for ChatGPT and Claude and MCP servers for AI agents. picrew_harness,243,"Catalog > Protocols, Tool Interfaces & Agent Contracts",AWS MCP Servers,https://github.com/awslabs/mcp,,,,,,repository,"AWS Labs suite of MCP servers exposing AWS documentation, cloud workflows, infrastructure operations, and service tooling to MCP clients." picrew_harness,244,"Catalog > Protocols, Tool Interfaces & Agent Contracts",Model Context Protocol,https://github.com/modelcontextprotocol/modelcontextprotocol,,,,,,repository,Core specification and docs for MCP-based tool and context interoperability. picrew_harness,245,"Catalog > Protocols, Tool Interfaces & Agent Contracts",MCP Registry,https://github.com/modelcontextprotocol/registry,,,,,,repository,Official MCP registry service and publisher tooling that lets clients discover MCP servers through a shared server catalog. picrew_harness,246,"Catalog > Protocols, Tool Interfaces & Agent Contracts",.NET Agent Skills,https://github.com/dotnet/skills,,,,,,repository,"Microsoft .NET team's curated plugins, skills, custom agents, and evaluation dashboard for grounding coding agents on .NET and C# workflows." picrew_harness,247,"Catalog > Protocols, Tool Interfaces & Agent Contracts",SkillHub,https://github.com/iflytek/skillhub,,,,,,repository,"Self-hosted enterprise agent skill registry with package publishing, versioning, discovery, namespaces, RBAC, reviews, and audit logs." picrew_harness,248,"Catalog > Protocols, Tool Interfaces & Agent Contracts",Builder.io Agent Skills,https://github.com/BuilderIO/skills,,,,,,repository,"Composable coding-agent skills for visual planning, visual recaps, watchdog reviews, plan arbitration, and autonomous execution discipline." picrew_harness,249,"Catalog > Protocols, Tool Interfaces & Agent Contracts",Agent Client Protocol,https://github.com/agentclientprotocol/agent-client-protocol,,,,,,repository,Open protocol that standardizes communication between code editors and coding agents. picrew_harness,250,"Catalog > Protocols, Tool Interfaces & Agent Contracts",directories (rules and MCP indexes),https://github.com/leerob/directories,,,,,,repository,Curated directories of agent rules and MCP servers for tool discovery. picrew_harness,251,"Catalog > Protocols, Tool Interfaces & Agent Contracts",Atmosphere,https://github.com/Atmosphere/atmosphere,,,,,,repository,"JVM runtime for streaming governable AI agents across MCP, A2A, AG-UI, and browser-facing transports." picrew_harness,252,"Catalog > Protocols, Tool Interfaces & Agent Contracts",Microsoft MCP Servers,https://github.com/microsoft/mcp,,,,,,repository,Microsoft's official MCP server catalog for enterprise data and tools. picrew_harness,253,"Catalog > Protocols, Tool Interfaces & Agent Contracts",LangChain MCP Adapters,https://github.com/langchain-ai/langchain-mcp-adapters,,,,,,repository,Adapters connecting LangChain components with MCP servers. picrew_harness,254,"Catalog > Protocols, Tool Interfaces & Agent Contracts",ACPX,https://github.com/openclaw/acpx,,,,,,repository,Headless CLI client for stateful Agent Client Protocol sessions. picrew_harness,255,"Catalog > Protocols, Tool Interfaces & Agent Contracts",Microsoft Agent Skills,https://github.com/microsoft/skills,,,,,,repository,"Microsoft-maintained skills, custom agents, AGENTS.md templates, and MCP configurations for Azure SDK and Microsoft AI Foundry coding-agent workflows." picrew_harness,256,"Catalog > Protocols, Tool Interfaces & Agent Contracts",GitAgentProtocol,https://github.com/open-gitagent/gitagent-protocol,,,,,,repository,"Git-native, framework-agnostic standard for defining agents, skills, workflows, tools, and runtime memory in repositories." picrew_harness,257,"Catalog > Protocols, Tool Interfaces & Agent Contracts",OpenGAP,https://github.com/open-gitagent/opengap,,,,,,repository,"Reference CLI for the Git Agent Protocol, turning repository files into portable agent manifests, rules, skills, workflows, tools, memory, hooks, and compliance contracts." picrew_harness,258,"Catalog > Protocols, Tool Interfaces & Agent Contracts",Microsoft Learn MCP,https://github.com/MicrosoftDocs/mcp,,,,,,repository,MCP server and CLI for grounding agents with Microsoft documentation sources. picrew_harness,259,"Catalog > Protocols, Tool Interfaces & Agent Contracts",IBM MCP,https://github.com/IBM/mcp,,,,,,repository,"IBM collection of MCP servers, clients, and developer tooling." picrew_harness,260,"Catalog > Protocols, Tool Interfaces & Agent Contracts",AGENT.md,https://github.com/agentmd/agent.md,,,,,,repository,Standardized machine-readable file format for agentic coding tools. picrew_harness,267,Catalog > Evaluation Harnesses & Benchmarks,Promptfoo,https://github.com/promptfoo/promptfoo,,,,,,repository,"Config-driven prompt/agent/RAG testing, comparison, and red-team evaluation tool." picrew_harness,268,Catalog > Evaluation Harnesses & Benchmarks,DeepEval,https://github.com/confident-ai/deepeval,,,,,,repository,LLM evaluation framework supporting agent and workflow quality testing. picrew_harness,269,Catalog > Evaluation Harnesses & Benchmarks,RAGAS,https://github.com/vibrantlabsai/ragas,,,,,,repository,Open evaluation toolkit for LLM and RAG quality metrics. picrew_harness,270,Catalog > Evaluation Harnesses & Benchmarks,lm-evaluation-harness,https://github.com/EleutherAI/lm-evaluation-harness,,,,,,repository,Popular benchmark harness for consistent LLM evaluation across tasks. picrew_harness,271,Catalog > Evaluation Harnesses & Benchmarks,Agent Reinforcement Trainer,https://github.com/OpenPipe/ART,,,,,,repository,"Agent reinforcement-training harness for multi-step agents with GRPO, trajectory capture, reward assignment, training loops, and LangGraph/MCP examples." picrew_harness,272,Catalog > Evaluation Harnesses & Benchmarks,SWE-bench,https://github.com/SWE-bench/SWE-bench,,,,,,repository,Standard benchmark for evaluating issue-fixing software engineering agents. picrew_harness,273,Catalog > Evaluation Harnesses & Benchmarks,Harbor,https://github.com/harbor-framework/harbor,,,,,,repository,Framework for running agent evaluations and constructing RL-style environments. picrew_harness,274,Catalog > Evaluation Harnesses & Benchmarks,LangWatch,https://github.com/langwatch/langwatch,,,,,,repository,"End-to-end platform for agent simulations, evaluation loops, and production testing." picrew_harness,275,Catalog > Evaluation Harnesses & Benchmarks,verifiers,https://github.com/PrimeIntellect-ai/verifiers,,,,,,repository,Library for RL environments and verifier-based evaluation loops. picrew_harness,276,Catalog > Evaluation Harnesses & Benchmarks,AgentBench,https://github.com/THUDM/AgentBench,,,,,,repository,Cross-environment benchmark for evaluating LLM agents as tool-using systems. picrew_harness,277,Catalog > Evaluation Harnesses & Benchmarks,EvalScope,https://github.com/modelscope/evalscope,,,,,,repository,Customizable framework for large-model benchmarking and performance evaluation. picrew_harness,278,Catalog > Evaluation Harnesses & Benchmarks,TEAQL Agent Kit,https://github.com/teaql/teaql-agent-kit,,,,,,repository,"Controlled evaluation environment for coding agents on auditable business software tasks with API adherence, framework discipline, self-repair, and human-gate metrics." picrew_harness,279,Catalog > Evaluation Harnesses & Benchmarks,tau2-bench,https://github.com/sierra-research/tau2-bench,,,,,,repository,Tool-agent-user interaction benchmark emphasizing multi-step execution quality. picrew_harness,280,Catalog > Evaluation Harnesses & Benchmarks,WebArena,https://github.com/web-arena-x/webarena,,,,,,repository,Self-hostable web environment and evaluation harness for autonomous web agents with reproducible end-to-end tasks. picrew_harness,281,Catalog > Evaluation Harnesses & Benchmarks,Meta-Harness,https://github.com/stanford-iris-lab/meta-harness,,,,,,repository,"Framework for automated search over task-specific model harnesses, with reference experiments for memory systems and terminal-agent scaffolds." picrew_harness,282,Catalog > Evaluation Harnesses & Benchmarks,NeMo Gym,https://github.com/NVIDIA-NeMo/Gym,,,,,,repository,Toolkit for building RL environments suitable for LLM/agent training and eval. picrew_harness,283,Catalog > Evaluation Harnesses & Benchmarks,ClawBench,https://github.com/TIGER-AI-Lab/ClawBench,,,,,,repository,"Browser-agent benchmark with live-site tasks, isolated containers, five-layer recording, and agentic scoring." picrew_harness,284,Catalog > Evaluation Harnesses & Benchmarks,TheAgentCompany,https://github.com/TheAgentCompany/TheAgentCompany,,,,,,repository,Agent benchmark with simulated software-company tasks for evaluating multi-step workplace autonomy. picrew_harness,285,Catalog > Evaluation Harnesses & Benchmarks,Claw-Eval,https://github.com/claw-eval/claw-eval,,,,,,repository,"Evaluation harness and benchmark for autonomous agents with human-verified tasks, trajectory auditing, and completion, safety, and robustness rubrics." picrew_harness,286,Catalog > Evaluation Harnesses & Benchmarks,Terminal-Bench,https://github.com/harbor-framework/terminal-bench,,,,,,repository,"Terminal-native benchmark suite for long-horizon, verification-heavy agent tasks." picrew_harness,287,Catalog > Evaluation Harnesses & Benchmarks,Inspect Evals,https://github.com/UKGovernmentBEIS/inspect_evals,,,,,,repository,Evaluation suite collection for Inspect AI workflows. picrew_harness,288,Catalog > Evaluation Harnesses & Benchmarks,auto-harness,https://github.com/neosigmaai/auto-harness,,,,,,repository,"Benchmark-gated optimization loop that mines failures, edits agent code, and guards against regressions overnight." picrew_harness,289,Catalog > Evaluation Harnesses & Benchmarks,SWE-Bench Pro,https://github.com/scaleapi/SWE-bench_Pro-os,,,,,,repository,Long-horizon software-engineering benchmark with reproducible Docker-based evaluation for issue-driven coding agents. picrew_harness,290,Catalog > Evaluation Harnesses & Benchmarks,WildClawBench,https://github.com/InternLM/WildClawBench,,,,,,repository,"In-the-wild benchmark that compares multiple agent harnesses on end-to-end multimodal, coding, safety, and productivity tasks inside a live OpenClaw environment." picrew_harness,291,Catalog > Evaluation Harnesses & Benchmarks,Agent Evaluation,https://github.com/awslabs/agent-evaluation,,,,,,repository,"AWS framework for testing virtual agents with evaluator-driven multi-turn conversations, hooks, and CI-friendly workflows." picrew_harness,292,Catalog > Evaluation Harnesses & Benchmarks,WorkArena,https://github.com/ServiceNow/WorkArena,,,,,,repository,Browser benchmark for practical enterprise-like knowledge work tasks. picrew_harness,293,Catalog > Evaluation Harnesses & Benchmarks,OpenHands Benchmarks,https://github.com/OpenHands/benchmarks,,,,,,repository,Evaluation harness and benchmark definitions for OpenHands systems. picrew_harness,294,Catalog > Evaluation Harnesses & Benchmarks,WebArena-Verified,https://github.com/ServiceNow/webarena-verified,,,,,,repository,Verified web-agent benchmark with deterministic evaluators. picrew_harness,295,Catalog > Evaluation Harnesses & Benchmarks,HarnessBench,https://github.com/reacher-z/HarnessBench,,,,,,repository,Benchmark for comparing agent harnesses on the same everyday web tasks with fixed models and per-harness containers. picrew_harness,302,Catalog > Observability & Reliability Operations,Langfuse,https://github.com/langfuse/langfuse,,,,,,repository,"Open-source LLM engineering platform for traces, metrics, prompts, and evals." picrew_harness,303,Catalog > Observability & Reliability Operations,MLflow,https://github.com/mlflow/mlflow,,,,,,repository,Broad AI engineering platform with monitoring and evaluation support for agents. picrew_harness,304,Catalog > Observability & Reliability Operations,Opik,https://github.com/comet-ml/opik,,,,,,repository,End-to-end debug/eval/monitoring stack for LLM apps and agent workflows. picrew_harness,305,Catalog > Observability & Reliability Operations,RagaAI Catalyst,https://github.com/raga-ai-hub/RagaAI-Catalyst,,,,,,repository,Agent observability and monitoring framework with timeline and graph analytics. picrew_harness,306,Catalog > Observability & Reliability Operations,TensorZero,https://github.com/tensorzero/tensorzero,,,,,,repository,"Open LLMOps stack unifying gateway, observability, evaluation, and optimization." picrew_harness,307,Catalog > Observability & Reliability Operations,Arize Phoenix,https://github.com/Arize-ai/phoenix,,,,,,repository,"Open platform for AI observability, tracing, and evaluation analytics." picrew_harness,308,Catalog > Observability & Reliability Operations,OpenLLMetry,https://github.com/traceloop/openllmetry,,,,,,repository,OpenTelemetry-based instrumentation for GenAI and LLM applications. picrew_harness,309,Catalog > Observability & Reliability Operations,Helicone,https://github.com/Helicone/helicone,,,,,,repository,Lightweight platform for monitoring and evaluating LLM traffic in production. picrew_harness,310,Catalog > Observability & Reliability Operations,AgentOps SDK,https://github.com/AgentOps-AI/agentops,,,,,,repository,Monitoring and benchmarking SDK for agent workflows with cost and trace tracking. picrew_harness,311,Catalog > Observability & Reliability Operations,Coze Loop,https://github.com/coze-dev/coze-loop,,,,,,repository,"Full-lifecycle agent optimization platform combining prompt development, automated evaluation, experiment management, trace collection, execution inspection, and production monitoring." picrew_harness,312,Catalog > Observability & Reliability Operations,Agenta,https://github.com/Agenta-AI/agenta,,,,,,repository,"Open-source LLMOps platform combining prompt management, evaluation, human and automated feedback, and tracing for production LLM applications." picrew_harness,313,Catalog > Observability & Reliability Operations,Latitude,https://github.com/latitude-dev/latitude-llm,,,,,,repository,Open-source agent engineering platform with eval and observability capabilities. picrew_harness,314,Catalog > Observability & Reliability Operations,TruLens,https://github.com/truera/trulens,,,,,,repository,"Evaluation and tracking platform for LLM apps and AI agents with traces, tool-call capture, and agentic evaluators." picrew_harness,315,Catalog > Observability & Reliability Operations,Laminar,https://github.com/lmnr-ai/lmnr,,,,,,repository,"Agent-focused observability stack with tracing, evaluation runs, monitoring, and dashboards." picrew_harness,316,Catalog > Observability & Reliability Operations,Desloppify,https://github.com/peteromallet/desloppify,,,,,,repository,"Agent-facing codebase quality harness with scans, scoring, LLM review, prioritized fix loops, persistent state, and CI gates." picrew_harness,317,Catalog > Observability & Reliability Operations,OpenLIT,https://github.com/openlit/openlit,,,,,,repository,"OpenTelemetry-native AI engineering platform for LLM and coding-agent observability, evaluations, guardrails, prompt management, secrets, and dashboards." picrew_harness,318,Catalog > Observability & Reliability Operations,claude-code-reverse,https://github.com/Yuyz0112/claude-code-reverse,,,,,,repository,Tooling to visualize and inspect Claude Code LLM interaction traces. picrew_harness,319,Catalog > Observability & Reliability Operations,Future AGI,https://github.com/future-agi/future-agi,,,,,,repository,"Self-hostable platform that closes the loop across agent tracing, evaluation, simulation, guardrails, and gateway operations." picrew_harness,320,Catalog > Observability & Reliability Operations,OpenInference,https://github.com/Arize-ai/openinference,,,,,,repository,Open instrumentation specification and tooling for AI observability. picrew_harness,321,Catalog > Observability & Reliability Operations,Judgeval,https://github.com/JudgmentLabs/judgeval,,,,,,repository,"Agent improvement SDK with OpenTelemetry tracing, agent judges, online monitoring, replay evaluations, CLI workflows, and MCP access to traces and behaviors." picrew_harness,322,Catalog > Observability & Reliability Operations,Agentic Harness Engineering,https://github.com/china-qijizhifeng/agentic-harness-engineering,,,,,,repository,"Observability-driven system for evolving coding-agent harness components through evaluate, analyze, and improve loops." picrew_harness,329,"Catalog > Guardrails, Security & Governance",OmniRoute,https://github.com/diegosouzapw/OmniRoute,,,,,,repository,"Local-first AI gateway for coding agents with quota-aware routing, provider fallback, credential pools, circuit breakers, usage controls, compression, MCP/A2A interfaces, and guardrails." picrew_harness,330,"Catalog > Guardrails, Security & Governance",LiteLLM,https://github.com/BerriAI/litellm,,,,,,repository,"Unified LLM gateway/proxy with cost tracking, load balancing, and guardrails." picrew_harness,331,"Catalog > Guardrails, Security & Governance",Kong,https://github.com/Kong/kong,,,,,,repository,API and AI gateway infrastructure useful for policy enforcement in agent systems. picrew_harness,332,"Catalog > Guardrails, Security & Governance",Parlant,https://github.com/emcie-co/parlant,,,,,,repository,"Interaction-control harness for customer-facing agents focused on consistent, predictable, and governed LLM behavior." picrew_harness,333,"Catalog > Guardrails, Security & Governance",SkillSpector,https://github.com/NVIDIA/SkillSpector,,,,,,repository,"Security scanner for AI agent skills with static and optional LLM analysis, vulnerability patterns, risk scoring, and CI-friendly report formats." picrew_harness,334,"Catalog > Guardrails, Security & Governance",Portkey Gateway,https://github.com/Portkey-AI/gateway,,,,,,repository,AI gateway with routing and guardrails for multi-model production traffic. picrew_harness,335,"Catalog > Guardrails, Security & Governance",CAI (Cybersecurity AI),https://github.com/aliasrobotics/cai,,,,,,repository,Security-focused agent framework for offensive/defensive AI workflows. picrew_harness,336,"Catalog > Guardrails, Security & Governance",Higress,https://github.com/higress-group/higress,,,,,,repository,"CNCF AI-native gateway for unified LLM API and MCP API management, including hosted remote MCP servers and gateway controls for agent tool access." picrew_harness,337,"Catalog > Guardrails, Security & Governance",Plano,https://github.com/katanemo/plano,,,,,,repository,"AI-native proxy and data plane with orchestration, safety, and observability." picrew_harness,338,"Catalog > Guardrails, Security & Governance",OpenAI Realtime Agents,https://github.com/openai/openai-realtime-agents,,,,,,repository,Advanced agentic realtime patterns with structured control and interaction loops. picrew_harness,339,"Catalog > Guardrails, Security & Governance",OpenAI CS Agents Demo,https://github.com/openai/openai-cs-agents-demo,,,,,,repository,Customer-service multi-agent demo highlighting handoffs and guardrail-like control points. picrew_harness,340,"Catalog > Guardrails, Security & Governance",Agent Governance Toolkit,https://github.com/microsoft/agent-governance-toolkit,,,,,,repository,"Runtime governance toolkit that deterministically enforces agent policy, identity, sandboxing, and audit controls before actions execute." picrew_harness,341,"Catalog > Guardrails, Security & Governance",AgentGateway,https://github.com/agentgateway/agentgateway,,,,,,repository,Agentic proxy gateway for AI agents and MCP server ecosystems. picrew_harness,342,"Catalog > Guardrails, Security & Governance",ContextForge,https://github.com/IBM/mcp-context-forge,,,,,,repository,"Registry and proxy layer that unifies MCP, A2A, and REST/gRPC endpoints with centralized governance and observability." picrew_harness,343,"Catalog > Guardrails, Security & Governance",Archestra,https://github.com/archestra-ai/archestra,,,,,,repository,"Enterprise AI platform with guardrails, MCP registry, and orchestration services." picrew_harness,344,"Catalog > Guardrails, Security & Governance",nono,https://github.com/always-further/nono,,,,,,repository,"Capability-based, policy-governed runtime that narrows agent access to explicit filesystem, network, credential, operation, sandbox, and audit capabilities." picrew_harness,345,"Catalog > Guardrails, Security & Governance",Tracecat,https://github.com/TracecatHQ/tracecat,,,,,,repository,AI automation platform for security teams with policy and workflow controls. picrew_harness,346,"Catalog > Guardrails, Security & Governance",Snyk Agent Scan,https://github.com/snyk/agent-scan,,,,,,repository,"Security scanner that inventories local agent harnesses, MCP servers, and skills, then checks for prompt injection, tool poisoning, malware payloads, and sensitive-data risks." picrew_harness,347,"Catalog > Guardrails, Security & Governance",Agent Vault,https://github.com/Infisical/agent-vault,,,,,,repository,"Credential proxy and vault that brokers agent API access without exposing real secrets, with egress filtering and request logging." picrew_harness,348,"Catalog > Guardrails, Security & Governance",ClawManager,https://github.com/Yuan-lab-LLM/ClawManager,,,,,,repository,"Kubernetes-native control plane for governing agent runtimes, AI gateway access, and reusable skills across multiple agent backends." picrew_harness,349,"Catalog > Guardrails, Security & Governance",Haft,https://github.com/m0n0x41d/haft,,,,,,repository,"Decision-governance harness that records falsifiable contracts, evidence, and commissions before agents execute." picrew_harness,350,"Catalog > Guardrails, Security & Governance",ClawKeeper,https://github.com/SafeAI-Lab-X/ClawKeeper,,,,,,repository,"Safety framework for autonomous agents combining skill-based policies, runtime plugins, and watcher-based governance." picrew_harness,351,"Catalog > Guardrails, Security & Governance",DefenseClaw,https://github.com/cisco-ai-defense/defenseclaw,,,,,,repository,"Security governance stack for OpenClaw and agentic runtimes, combining admission control, runtime inspection, CodeGuard checks, sandbox policy, audit logs, and observability export." picrew_harness,352,"Catalog > Guardrails, Security & Governance",Cordum,https://github.com/cordum-io/cordum,,,,,,repository,"Agent control plane for pre-execution policies, approval gates, safety-kernel verdicts, live monitoring, audit trails, and Claude Code compliance firewall hooks." picrew_harness,353,"Catalog > Guardrails, Security & Governance",Sponsio,https://github.com/SponsioLabs/Sponsio,,,,,,repository,Runtime enforcement layer that checks every agent action against deterministic contracts before execution. picrew_harness,354,"Catalog > Guardrails, Security & Governance",DashClaw,https://github.com/ucsandman/DashClaw,,,,,,repository,"Governance layer that intercepts risky agent actions, enforces policy, routes approvals, and records audit-ready decision trails." picrew_harness,355,"Catalog > Guardrails, Security & Governance",Tandem,https://github.com/frumu-ai/tandem,,,,,,repository,"Governed runtime authority layer for agents with scoped execution, tool visibility, permissioned memory, approval gates, and audit trails." picrew_harness,362,Catalog > Reference Harness Implementations,OpenClaw,https://github.com/openclaw/openclaw,,,,,,repository,"Local-first personal assistant harness with a gateway control plane for sessions, channels, tools, events, skills, and sandboxed non-main agents." picrew_harness,363,Catalog > Reference Harness Implementations,Hermes Agent,https://github.com/NousResearch/hermes-agent,,,,,,repository,"Self-improving agent runtime with memory, skill creation, subagents, scheduled automations, and pluggable terminal backends." picrew_harness,364,Catalog > Reference Harness Implementations,DeepSeek Harness,https://github.com/deepseek-ai/deepseek-harness,,,,,,repository,"DeepSeek's official plugin-first agent harness with replaceable agent loops, model adapters, tool registries, durable session logs, sandbox and approval policies, and web or headless profiles." picrew_harness,365,Catalog > Reference Harness Implementations,OpenCode,https://github.com/anomalyco/opencode,,,,,,repository,"Open-source coding agent with built-in plan/build roles, subagents, LSP support, and a client-server runtime." picrew_harness,366,Catalog > Reference Harness Implementations,Claw Code,https://github.com/ultraworkers/claw-code,,,,,,repository,"Public Rust implementation of the claw CLI agent harness with auth, sessions, parity checks, container workflows, and terminal execution guidance." picrew_harness,367,Catalog > Reference Harness Implementations,Claude Code,https://github.com/anthropics/claude-code,,,,,,repository,"Official terminal coding agent that understands codebases and executes editing, debugging, and Git workflows through natural language." picrew_harness,368,Catalog > Reference Harness Implementations,Codex CLI,https://github.com/openai/codex,,,,,,repository,Terminal-native coding agent that runs locally and exposes practical agent workflows for software tasks. picrew_harness,369,Catalog > Reference Harness Implementations,Browser Use,https://github.com/browser-use/browser-use,,,,,,repository,"Browser-agent framework that exposes websites to LLMs through browser state, tools, cloud browsers, and benchmarked task runs." picrew_harness,371,Catalog > Reference Harness Implementations,Gemini CLI,https://github.com/google-gemini/gemini-cli,,,,,,repository,"Open-source terminal agent with built-in tools, MCP support, checkpointing, and sandboxing controls." picrew_harness,372,Catalog > Reference Harness Implementations,OpenDesign,https://github.com/nexu-io/open-design,,,,,,repository,"Local-first design-agent workspace with declarative CLI runtime adapters, structured event streaming, session resume, cancellation, composable skills, and sandboxed artifact previews." picrew_harness,373,Catalog > Reference Harness Implementations,OpenHands,https://github.com/OpenHands/OpenHands,,,,,,repository,Open-source AI software engineer focused on repo-level coding task execution. picrew_harness,374,Catalog > Reference Harness Implementations,LobeHub,https://github.com/lobehub/lobehub,,,,,,repository,"Chief-agent-operator platform for scheduling, running, and reporting on multi-agent workstreams." picrew_harness,375,Catalog > Reference Harness Implementations,Paperclip,https://github.com/paperclipai/paperclip,,,,,,repository,"Managed-agent control plane with org charts, ticketing, budgets, heartbeats, and audit trails for coordinating agent teams." picrew_harness,376,Catalog > Reference Harness Implementations,learn-claude-code,https://github.com/shareAI-lab/learn-claude-code,,,,,,repository,Hands-on harness tutorial for building Claude Code-like systems from scratch. picrew_harness,377,Catalog > Reference Harness Implementations,Cline,https://github.com/cline/cline,,,,,,repository,"Open-source coding agent spanning IDE, terminal, SDK, and kanban surfaces with shared approvals, MCP, checkpoints, and agent teams." picrew_harness,378,Catalog > Reference Harness Implementations,Open Interpreter,https://github.com/openinterpreter/openinterpreter,,,,,,repository,"Terminal coding agent with selectable harness emulation, native sandboxing, provider switching, skills, hooks, permissions, MCP, and AGENTS.md support." picrew_harness,379,Catalog > Reference Harness Implementations,OpenManus,https://github.com/FoundationAgents/OpenManus,,,,,,repository,Open foundation for broad autonomous agent workflows with coding-heavy use cases. picrew_harness,380,Catalog > Reference Harness Implementations,goose,https://github.com/aaif-goose/goose,,,,,,repository,"Linux Foundation AAIF open-source AI agent spanning desktop, CLI, and embeddable API with provider adapters, ACP access, and MCP extensions." picrew_harness,381,Catalog > Reference Harness Implementations,Multica,https://github.com/multica-ai/multica/,,,,,,repository,"Managed-agents platform that assigns issues to coding agents, routes execution through runtimes, and compounds reusable skills." picrew_harness,382,Catalog > Reference Harness Implementations,CLI-Anything,https://github.com/HKUDS/CLI-Anything,,,,,,repository,CLI agent system that unifies command-line tool usage in agent loops. picrew_harness,383,Catalog > Reference Harness Implementations,aider,https://github.com/Aider-AI/aider,,,,,,repository,"Terminal coding assistant with repo mapping, git-aware edits, and built-in lint/test feedback loops." picrew_harness,384,Catalog > Reference Harness Implementations,nanobot,https://github.com/HKUDS/nanobot,,,,,,repository,"Ultra-lightweight agent runtime with WebUI, chat channels, tools, memory, MCP, model routing, deployment, and long-running goal support." picrew_harness,385,Catalog > Reference Harness Implementations,CowAgent,https://github.com/zhayujie/CowAgent,,,,,,repository,"Reference agent harness implementation with planning, memory, knowledge, skills, tools, MCP integration, schedulers, browser automation, and multi-channel delivery." picrew_harness,386,Catalog > Reference Harness Implementations,CodeWhale,https://github.com/Hmbown/CodeWhale,,,,,,repository,"Local-first coding-agent harness with explicit authority ordering, evidence loops, approval-gated tools, side-git snapshots, rollback, subagents, and runtime APIs." picrew_harness,387,Catalog > Reference Harness Implementations,OpenHuman,https://github.com/tinyhumansai/openhuman,,,,,,repository,"Local-first agent harness combining persistent memory, checkpointed graph orchestration, approval-gated workflows, failure circuit breakers, and replayable event journals." picrew_harness,388,Catalog > Reference Harness Implementations,Claude Code Plugins: Orchestration and Automation,https://github.com/wshobson/agents,,,,,,repository,"Production-ready Claude Code plugin marketplace bundling agents, skills, tools, and multi-agent workflow orchestrators." picrew_harness,389,Catalog > Reference Harness Implementations,oh-my-claudecode,https://github.com/Yeachan-Heo/oh-my-claudecode,,,,,,repository,"Team-first orchestration layer for Claude Code with staged multi-agent execution, worktree-aware setup, and persistent session artifacts." picrew_harness,390,Catalog > Reference Harness Implementations,Agent TARS,https://github.com/bytedance/UI-TARS-desktop,,,,,,repository,"Multimodal computer and browser agent stack with CLI/Web UI, hybrid GUI/DOM browser control, MCP tools, event streams, and sandbox support." picrew_harness,391,Catalog > Reference Harness Implementations,OpenCodeReview,https://github.com/alibaba/open-code-review,,,,,,repository,"Production-tested code-review agent harness combining deterministic file coverage and rule selection with isolated subagents, tool-driven context retrieval, comment verification, sessions, CI, and tracing." picrew_harness,392,Catalog > Reference Harness Implementations,DeepSeek-Reasonix,https://github.com/esengine/DeepSeek-Reasonix,,,,,,repository,"DeepSeek-native coding-agent harness optimized for prefix-cache stability, with separate planner and executor sessions, cache-aware compaction, MCP plugins, permissions, workspace sandboxing, and per-turn checkpoints." picrew_harness,393,Catalog > Reference Harness Implementations,oh-my-codex,https://github.com/Yeachan-Heo/oh-my-codex,,,,,,repository,"Workflow layer for OpenAI Codex CLI with stronger session startup, standard planning-to-completion flows, durable state, skills, hooks, and worktree launches." picrew_harness,394,Catalog > Reference Harness Implementations,ZeroClaw,https://github.com/zeroclaw-labs/zeroclaw,,,,,,repository,"Single-binary agent runtime with providers, channels, tools, memory, SOPs, approval gates, sandboxing, ACP, and tool receipts." picrew_harness,395,Catalog > Reference Harness Implementations,oh-my-pi,https://github.com/can1357/oh-my-pi,,,,,,repository,"Terminal AI coding agent with edit safety, LSP integration, and subagent support." picrew_harness,396,Catalog > Reference Harness Implementations,NanoClaw,https://github.com/nanocoai/nanoclaw,,,,,,repository,"Container-isolated Claude agent harness with channel routing, scheduled jobs, per-group memory, and small-codebase customization." picrew_harness,397,Catalog > Reference Harness Implementations,Crush,https://github.com/charmbracelet/crush,,,,,,repository,"Terminal coding agent that wires project tools, code, workflows, sessions, LSP context, model switching, and MCP extensions into an LLM runtime." picrew_harness,398,Catalog > Reference Harness Implementations,Vibe Kanban,https://github.com/BloopAI/vibe-kanban,,,,,,repository,"Kanban control plane for planning, running, reviewing, and merging work from coding agents in isolated workspaces." picrew_harness,399,Catalog > Reference Harness Implementations,Qwen Code,https://github.com/QwenLM/qwen-code,,,,,,repository,Terminal-native open-source coding agent tuned for practical dev loops. picrew_harness,400,Catalog > Reference Harness Implementations,Kilo Code,https://github.com/Kilo-Org/kilocode,,,,,,repository,"Open-source coding agent spanning VS Code, JetBrains, CLI, cloud agents, reviews, custom modes, browser and terminal control, MCP marketplace, and autonomous CI runs." picrew_harness,401,Catalog > Reference Harness Implementations,cmux,https://github.com/manaflow-ai/cmux,,,,,,repository,"Native macOS terminal and browser workspace for AI coding agents with notifications, split panes, and scriptable control." picrew_harness,402,Catalog > Reference Harness Implementations,Grok Build,https://github.com/xai-org/grok-build,,,,,,repository,"Official Rust coding-agent harness and TUI with interactive, headless, and ACP modes plus MCP, skills, plugins, hooks, sandboxing, subagents, and long-running task support." picrew_harness,403,Catalog > Reference Harness Implementations,Compound Engineering,https://github.com/EveryInc/compound-engineering-plugin,,,,,,repository,"Cross-agent engineering plugin that codifies brainstorming, planning, worktree execution, review, and knowledge compounding loops." picrew_harness,404,Catalog > Reference Harness Implementations,SuperClaude Framework,https://github.com/SuperClaude-Org/SuperClaude_Framework,,,,,,repository,"Configuration framework adding commands, personas, and method templates to coding agents." picrew_harness,405,Catalog > Reference Harness Implementations,Prime Agent,https://github.com/PrimeIntellect-ai/prime-agent,,,,,,repository,"Long-running coding and research harness with a persistent Python REPL, recursive subagents, resumable daemon sessions, durable harness state, and budgeted autonomous quality gates." picrew_harness,406,Catalog > Reference Harness Implementations,SWE-agent,https://github.com/SWE-agent/SWE-agent,,,,,,repository,Research-grade coding agent that resolves GitHub issues with explicit tooling loops. picrew_harness,407,Catalog > Reference Harness Implementations,jcode,https://github.com/1jehuang/jcode,,,,,,repository,"Rust coding-agent harness built for multi-session workflows, customization, memory, and terminal performance." picrew_harness,408,Catalog > Reference Harness Implementations,Devika,https://github.com/stitionai/devika,,,,,,repository,Open-source coding assistant system for planning and implementing development tasks. picrew_harness,409,Catalog > Reference Harness Implementations,OpenFang,https://github.com/RightNow-AI/openfang,,,,,,repository,"Rust agent operating system with autonomous capability packages, manifests, guardrails, tools, memory, sandboxing, audit trails, and channel adapters." picrew_harness,410,Catalog > Reference Harness Implementations,Paseo,https://github.com/getpaseo/paseo,,,,,,repository,"Multi-device coding-agent daemon and client stack for orchestrating local agents, parallel runs, and cross-provider workflows." picrew_harness,411,Catalog > Reference Harness Implementations,DeepCode,https://github.com/HKUDS/DeepCode,,,,,,repository,"Coding-agent harness with durable sessions, goal-driven loops, evidence-based completion, permission and sandbox controls, reusable skills, parallel agents, context compaction, and headless automation." picrew_harness,412,Catalog > Reference Harness Implementations,OpenHarness,https://github.com/HKUDS/OpenHarness,,,,,,repository,"Open agent harness implementation covering tool use, skills, memory, permissions, and multi-agent coordination." picrew_harness,413,Catalog > Reference Harness Implementations,Learn Harness Engineering,https://github.com/walkinglabs/learn-harness-engineering,,,,,,repository,"Project-based harness engineering course with runnable templates and staged implementations for instructions, state, verification, observability, automated loops, graph workflows, rollback, and human approval." picrew_harness,414,Catalog > Reference Harness Implementations,Eigent,https://github.com/eigent-ai/eigent,,,,,,repository,Open-source desktop cowork agent for autonomous task execution and productivity. picrew_harness,415,Catalog > Reference Harness Implementations,Aperant,https://github.com/AndyMik90/Aperant,,,,,,repository,"Autonomous multi-agent coding framework with parallel execution, isolated workspaces, QA loops, and persistent memory." picrew_harness,416,Catalog > Reference Harness Implementations,Superset,https://github.com/superset-sh/superset,,,,,,repository,Worktree-based desktop orchestrator for running and reviewing parallel CLI coding agents from one workspace. picrew_harness,417,Catalog > Reference Harness Implementations,IronClaw,https://github.com/nearai/ironclaw,,,,,,repository,"Security-first personal agent harness with WASM sandboxing, routines, tool plugins, and persistent memory." picrew_harness,418,Catalog > Reference Harness Implementations,Agent S,https://github.com/simular-ai/Agent-S,,,,,,repository,"Open-source computer-use agent framework with grounding models, reflection, local code execution option, and OSWorld-style evaluation support." picrew_harness,419,Catalog > Reference Harness Implementations,Agent Orchestrator,https://github.com/Untrivial-ai/agent-orchestrator,,,,,,repository,Worktree-based orchestration layer for parallel coding agents with autonomous CI and review feedback handling. picrew_harness,420,Catalog > Reference Harness Implementations,holaOS,https://github.com/holaboss-ai/holaOS,,,,,,repository,"Desktop-first long-horizon agent environment with runtime, memory, tools, apps, and durable state." picrew_harness,421,Catalog > Reference Harness Implementations,GitHub Copilot CLI,https://github.com/github/copilot-cli,,,,,,repository,"Official terminal coding agent built on GitHub's Copilot harness with MCP extensibility, approval controls, and GitHub-native context." picrew_harness,422,Catalog > Reference Harness Implementations,ML Intern,https://github.com/huggingface/ml-intern,,,,,,repository,"Hugging Face ML engineering agent with local and sandbox tool runtimes, private Space sandboxes, trace logging, and CLI execution." picrew_harness,423,Catalog > Reference Harness Implementations,Open SWE,https://github.com/langchain-ai/open-swe,,,,,,repository,Asynchronous open-source coding agent focused on software issue workflows. picrew_harness,424,Catalog > Reference Harness Implementations,GitHub Copilot SDK,https://github.com/github/copilot-sdk,,,,,,repository,"Official multi-language SDK for embedding the Copilot CLI agent runtime with sessions, planning, tools, file edits, permissions, custom agents, skills, MCP, hooks, and JSON-RPC lifecycle management." picrew_harness,425,Catalog > Reference Harness Implementations,Harness,https://github.com/revfactory/harness,,,,,,repository,"Claude Code meta-factory that generates domain-specific agent teams, skills, orchestration patterns, and validation steps from a project description." picrew_harness,426,Catalog > Reference Harness Implementations,OSAURUS,https://github.com/osaurus-ai/osaurus,,,,,,repository,Native macOS harness for autonomous coding agents with persistent memory. picrew_harness,427,Catalog > Reference Harness Implementations,mini-swe-agent,https://github.com/SWE-agent/mini-swe-agent,,,,,,repository,Minimal coding agent implementation with strong benchmark competitiveness. picrew_harness,428,Catalog > Reference Harness Implementations,Munder Difflin,https://github.com/chaitanyagiri/munder-difflin,,,,,,repository,"Desktop multi-agent harness that wraps terminal-agent CLIs with hive mailboxes, shared memory, an orchestrator, approvals, worktrees, and telemetry." picrew_harness,429,Catalog > Reference Harness Implementations,Yuxi,https://github.com/xerrors/Yuxi,,,,,,repository,"Multi-tenant agent harness platform combining RAG, knowledge graphs, LangGraph orchestration, skills, MCP, subagents, and sandbox tools." picrew_harness,430,Catalog > Reference Harness Implementations,Webwright,https://github.com/microsoft/Webwright,,,,,,repository,Minimal browser-agent harness that lets coding models solve web tasks by writing rerunnable Playwright scripts in a workspace. picrew_harness,431,Catalog > Reference Harness Implementations,Google Agents CLI,https://github.com/google/agents-cli,,,,,,repository,"Google Cloud CLI and skill bundle that gives coding agents scaffold, evaluation, deployment, publishing, and observability workflows." picrew_harness,432,Catalog > Reference Harness Implementations,AgentTeams,https://github.com/agentscope-ai/AgentTeams,,,,,,repository,"Collaborative multi-agent OS with manager-worker coordination, shared state, human-in-the-loop oversight, pluggable runtimes, and Kubernetes deployment through Matrix rooms." picrew_harness,433,Catalog > Reference Harness Implementations,1Code,https://github.com/21st-dev/1code,,,,,,repository,"Desktop-first coding-agent orchestrator with worktree isolation, background sandboxes, MCP tooling, and automation triggers." picrew_harness,434,Catalog > Reference Harness Implementations,AI-DLC Workflows,https://github.com/awslabs/aidlc-workflows,,,,,,repository,"Official AWS workflow ruleset that steers coding agents through adaptive phases, quality gates, and IDE-specific context files." picrew_harness,435,Catalog > Reference Harness Implementations,gptme,https://github.com/gptme/gptme,,,,,,repository,"Terminal-native personal agent with local tools, shell and web access, provider-agnostic models, plugins, skills, MCP, guardrails, and autonomous loops." picrew_harness,436,Catalog > Reference Harness Implementations,TinyAGI,https://github.com/TinyAGI/tinyagi,,,,,,repository,Team-style agent orchestrator for one-person-company style autonomous workflows. picrew_harness,437,Catalog > Reference Harness Implementations,Open Claude Cowork,https://github.com/DevAgentForge/Open-Claude-Cowork,,,,,,repository,Desktop coding cowork assistant that turns agent orchestration into GUI workflows. picrew_harness,438,Catalog > Reference Harness Implementations,Amazon Bedrock AgentCore Samples,https://github.com/awslabs/agentcore-samples,,,,,,repository,"Official sample suite for deploying and operating agents with runtime, gateway, memory, observability, evaluation, and policy layers." picrew_harness,439,Catalog > Reference Harness Implementations,Maestro,https://github.com/RunMaestro/Maestro,,,,,,repository,"Desktop command center for parallel coding agents with worktree isolation, queued tasks, auto-run playbooks, and reusable sessions." picrew_harness,440,Catalog > Reference Harness Implementations,Claude Code Harness,https://github.com/Chachamaru127/claude-code-harness,,,,,,repository,"Claude Code development harness that turns agent work into a repeatable spec, plan, work, review, and release loop." picrew_harness,441,Catalog > Reference Harness Implementations,Kimchi,https://github.com/getkimchi/kimchi,,,,,,repository,"Terminal coding-agent harness with multi-model orchestration roles, phase tracking, persistent project state, and subagent delegation." picrew_harness,442,Catalog > Reference Harness Implementations,Open Cowork,https://github.com/OpenCoworkAI/open-cowork,,,,,,repository,"Desktop agent app with VM-backed sandboxing, MCP connectors, GUI control, and built-in skill workflows." picrew_harness,443,Catalog > Reference Harness Implementations,Pilot Shell,https://github.com/maxritter/pilot-shell,,,,,,repository,"Claude Code and Codex workflow harness with spec-driven planning, enforced TDD, persistent memory, quality gates, and reusable skills." picrew_harness,444,Catalog > Reference Harness Implementations,thClaws,https://github.com/thClaws/thClaws,,,,,,repository,"Native Rust agent workspace with shared agent loop, sessions, tools, skills, MCP, memory, hooks, and sandboxing." picrew_harness,445,Catalog > Reference Harness Implementations,mini-coding-agent,https://github.com/rasbt/mini-coding-agent,,,,,,repository,"Minimal coding agent harness illustrating approvals, memory, bounded delegation, and durable transcripts." picrew_harness,446,Catalog > Reference Harness Implementations,MateClaw,https://github.com/matevip/mateclaw,,,,,,repository,"Self-hosted multi-user agent harness with StateGraph reasoning, skills, MCP/ACP registry, approvals, audit trail, and channel adapters." picrew_harness,447,Catalog > Reference Harness Implementations,Flock,https://github.com/Onelevenvy/flock,,,,,,repository,"Desktop multi-agent harness with visual workflows, local and sandboxed execution, tool approvals, VNC takeover, MCP, skills, and scheduled tasks." picrew_harness,448,Catalog > Reference Harness Implementations,codex-autorunner,https://github.com/Git-on-my-level/codex-autorunner,,,,,,repository,"Meta-harness that treats tickets as the control plane for long-running coding agents, with queue execution, hub UI, and chat notifications." picrew_harness,449,Catalog > Reference Harness Implementations,CUGA,https://github.com/cuga-project/cuga-agent,,,,,,repository,"Enterprise generalist agent harness with MCP/OpenAPI tools, policy controls, HITL gates, memory, skills, and hybrid API/browser execution." picrew_harness,450,Catalog > Reference Harness Implementations,CheetahClaws,https://github.com/SafeRL-Lab/cheetahclaws,,,,,,repository,"Python agent harness infrastructure for long-horizon, multi-model, tool-using coding assistants with MCP, skills, memory, approvals, checkpoints, and bridges." picrew_harness,451,Catalog > Reference Harness Implementations,Dexto,https://github.com/truffle-ai/dexto,,,,,,repository,"Open agent harness for AI applications with YAML configs, stateful sessions, tool orchestration, memory, observability, permissions, and subagents." picrew_harness,452,Catalog > Reference Harness Implementations,OpenClaw.NET,https://github.com/clawdotnet/openclaw.net,,,,,,repository,"NativeAOT-friendly .NET agent runtime and gateway with tools, memory, MCP, governance ledger, evidence bundles, and harness regression tests." picrew_harness,453,Catalog > Reference Harness Implementations,Utah,https://github.com/inngest/utah,,,,,,repository,"Inngest-powered durable agent harness with a think-act-observe loop, step-level retries, singleton concurrency, cancellation, and multi-channel adapters." picrew_harness,460,Catalog > Essential Readings & Ecosystem Maps,awesome-claude-code,https://github.com/hesreallyhim/awesome-claude-code,,,,,,repository,"Community collection of Claude Code skills, hooks, and orchestrator tooling." picrew_harness,461,Catalog > Essential Readings & Ecosystem Maps,Awesome Agent Skills,https://github.com/VoltAgent/awesome-agent-skills,,,,,,repository,"Curated cross-harness map of official and community Agent Skills for Claude Code, Codex, Gemini CLI, Cursor, OpenCode, Copilot, and related hosts." picrew_harness,462,Catalog > Essential Readings & Ecosystem Maps,awesome-agentic-patterns,https://github.com/nibzard/awesome-agentic-patterns,,,,,,repository,Catalog of reusable agentic design patterns and implementation motifs. picrew_harness,463,Catalog > Essential Readings & Ecosystem Maps,awesome-mcp-servers,https://github.com/wong2/awesome-mcp-servers,,,,,,repository,Curated MCP server index for tool interoperability in agent systems. picrew_harness,464,Catalog > Essential Readings & Ecosystem Maps,awesome-harness-engineering,https://github.com/walkinglabs/awesome-harness-engineering,,,,,,repository,"Curated list focused on harness engineering articles, benchmarks, and implementations." picrew_harness,465,Catalog > Essential Readings & Ecosystem Maps,12 Factor Agents,https://www.humanlayer.dev/blog/12-factor-agents,,,,,,grey,Operations-oriented principles for building maintainable production agents. picrew_harness,466,Catalog > Essential Readings & Ecosystem Maps,"Agent Frameworks, Runtimes, and Harnesses, oh my!",https://blog.langchain.com/agent-frameworks-runtimes-and-harnesses-oh-my/,,,,,,grey,Clear decomposition of framework vs runtime vs harness responsibilities. picrew_harness,467,Catalog > Essential Readings & Ecosystem Maps,An open-source spec for Codex orchestration: Symphony.,https://openai.com/index/open-source-codex-orchestration-symphony/,,,,,,grey,OpenAI's orchestration write-up on turning issue trackers into always-on control planes for coding agents. picrew_harness,468,Catalog > Essential Readings & Ecosystem Maps,"Building a safe, effective sandbox to enable Codex on Windows",https://openai.com/index/building-codex-windows-sandbox/,,,,,,grey,"OpenAI's implementation report on enforcing filesystem and network boundaries for Codex on Windows with restricted tokens, SIDs, and a dedicated command runner." picrew_harness,469,Catalog > Essential Readings & Ecosystem Maps,Building Effective AI Agents,https://www.anthropic.com/engineering/building-effective-agents,,,,,,grey,Anthropic's practical guidance on when to use workflows vs. autonomous agents and how to structure them. picrew_harness,470,Catalog > Essential Readings & Ecosystem Maps,Claude Agent SDK overview,https://code.claude.com/docs/en/agent-sdk/overview,,,,,,grey,"Official Claude Agent SDK documentation for building agents on the Claude Code harness, including tools, MCP, permissions, hooks, subagents, and session flows." picrew_harness,471,Catalog > Essential Readings & Ecosystem Maps,Claude Code auto mode,https://www.anthropic.com/engineering/claude-code-auto-mode,,,,,,grey,Anthropic's write-up on classifier-backed approval delegation for safer high-autonomy coding-agent runs. picrew_harness,472,Catalog > Essential Readings & Ecosystem Maps,Code execution with MCP,https://www.anthropic.com/engineering/code-execution-with-mcp,,,,,,grey,Anthropic's design notes on controlled code execution via MCP boundaries. picrew_harness,473,Catalog > Essential Readings & Ecosystem Maps,Demystifying Evals for AI Agents,https://www.anthropic.com/engineering/demystifying-evals-for-ai-agents,,,,,,grey,Methodology for designing robust agent evals in non-deterministic trajectories. picrew_harness,474,Catalog > Essential Readings & Ecosystem Maps,Effective context engineering for AI agents,https://www.anthropic.com/engineering/effective-context-engineering-for-ai-agents,,,,,,grey,Guidance on context-window budgeting and working-state management for agents. picrew_harness,475,Catalog > Essential Readings & Ecosystem Maps,Effective harnesses for long-running agents,https://www.anthropic.com/engineering/effective-harnesses-for-long-running-agents,,,,,,grey,"Practical guide to maintaining state, resumability, and reliability over long agent runs." picrew_harness,476,Catalog > Essential Readings & Ecosystem Maps,Evaluating Deep Agents: Our Learnings,https://blog.langchain.com/evaluating-deep-agents-our-learnings/,,,,,,grey,LangChain's practical lessons on evaluating stateful and long-horizon agents. picrew_harness,477,Catalog > Essential Readings & Ecosystem Maps,From model to agent: Equipping the Responses API with a computer environment,https://openai.com/index/equip-responses-api-computer-environment/,,,,,,grey,"OpenAI's engineering guide to agent execution with the Responses API, hosted shell containers, orchestration loops, compaction, skills, persistent files, and restricted networking." picrew_harness,478,Catalog > Essential Readings & Ecosystem Maps,Harness design for long-running application development,https://www.anthropic.com/engineering/harness-design-long-running-apps,,,,,,grey,Follow-up article on improving long-running app generation through harness structure. picrew_harness,479,Catalog > Essential Readings & Ecosystem Maps,Harness Engineering (Martin Fowler),https://martinfowler.com/articles/exploring-gen-ai/harness-engineering.html,,,,,,grey,Architectural perspective on harness engineering and entropy control. picrew_harness,480,Catalog > Essential Readings & Ecosystem Maps,Harness engineering (OpenAI),https://openai.com/index/harness-engineering/,,,,,,grey,Field report on building reliable agent-first software via harness constraints and verification. picrew_harness,481,Catalog > Essential Readings & Ecosystem Maps,How GPT-5.6 fuses frontier intelligence with frontier efficiency,https://openai.com/index/gpt-5-6-frontier-intelligence-efficiency/,,,,,,grey,"OpenAI's engineering account of the Rust agentic harness behind Codex and ChatGPT Work, covering deferred tool discovery, bounded outputs, deterministic tool ordering, and prompt-cache preservation." picrew_harness,482,Catalog > Essential Readings & Ecosystem Maps,How we built our multi-agent research system,https://www.anthropic.com/engineering/multi-agent-research-system,,,,,,grey,Anthropic architecture write-up on role separation and coordination in multi-agent systems. picrew_harness,483,Catalog > Essential Readings & Ecosystem Maps,How we contain Claude across products,https://www.anthropic.com/engineering/how-we-contain-claude,,,,,,grey,"Anthropic's cross-product containment architecture for claude.ai, Claude Code, and Cowork, spanning sandboxes, VMs, filesystem boundaries, egress controls, and approval fatigue." picrew_harness,484,Catalog > Essential Readings & Ecosystem Maps,Improving Deep Agents with harness engineering,https://blog.langchain.com/improving-deep-agents-with-harness-engineering/,,,,,,grey,Evidence that harness improvements alone can move benchmark performance. picrew_harness,485,Catalog > Essential Readings & Ecosystem Maps,"Introducing Agent Executor, Google’s distributed Agent Runtime",https://cloud.google.com/blog/products/ai-machine-learning/agent-executor-googles-distributed-agent-runtime/,,,,,,grey,"Google's distributed agent runtime design for durable execution, event-log snapshots, secure isolation, session consistency, reconnection, and trajectory branching." picrew_harness,486,Catalog > Essential Readings & Ecosystem Maps,Making Claude Code more secure and autonomous with sandboxing,https://www.anthropic.com/engineering/claude-code-sandboxing,,,,,,grey,How Anthropic uses sandbox boundaries to raise agent autonomy without giving up security controls. picrew_harness,487,Catalog > Essential Readings & Ecosystem Maps,Quantifying infrastructure noise in agentic coding evals,https://www.anthropic.com/engineering/infrastructure-noise,,,,,,grey,Analysis of how infrastructure choices impact coding-agent benchmark outcomes. picrew_harness,488,Catalog > Essential Readings & Ecosystem Maps,Running Codex safely at OpenAI,https://openai.com/index/running-codex-safely/,,,,,,grey,"OpenAI's operational blueprint for governing Codex with sandbox and approval policies, network controls, managed identity, configuration, telemetry, and audit trails." picrew_harness,489,Catalog > Essential Readings & Ecosystem Maps,Scaling Managed Agents: Decoupling the brain from the hands,https://www.anthropic.com/engineering/managed-agents,,,,,,grey,"Anthropic's meta-harness architecture for decoupling session logs, harness loops, and sandboxes in long-horizon agents." picrew_harness,490,Catalog > Essential Readings & Ecosystem Maps,Skill Issue: Harness Engineering for Coding Agents,https://www.humanlayer.dev/blog/skill-issue-harness-engineering-for-coding-agents,,,,,,grey,Practical breakdown of why coding-agent quality depends heavily on harness setup. picrew_harness,491,Catalog > Essential Readings & Ecosystem Maps,Testing Agent Skills Systematically with Evals,https://developers.openai.com/blog/eval-skills,,,,,,grey,OpenAI Developers guide for turning agent traces into repeatable skill evaluations. picrew_harness,492,Catalog > Essential Readings & Ecosystem Maps,The Anatomy of an Agent Harness,https://blog.langchain.com/the-anatomy-of-an-agent-harness/,,,,,,grey,Conceptual decomposition of agent harness components and their responsibilities. picrew_harness,493,Catalog > Essential Readings & Ecosystem Maps,The next evolution of the Agents SDK,https://openai.com/index/the-next-evolution-of-the-agents-sdk/,,,,,,grey,"OpenAI's product and engineering post on model-native agent harnesses, native sandbox execution, manifests, memory, and filesystem and shell tools." picrew_harness,494,Catalog > Essential Readings & Ecosystem Maps,Unlocking the Codex harness: how we built the App Server,https://openai.com/index/unlocking-the-codex-harness/,,,,,,grey,"OpenAI's architecture deep dive into sharing the Codex harness across clients through App Server, thread persistence, JSON-RPC, sandboxed tools, MCP, skills, and reconnectable sessions." picrew_harness,495,Catalog > Essential Readings & Ecosystem Maps,Unrolling the Codex agent loop,https://openai.com/index/unrolling-the-codex-agent-loop/,,,,,,grey,"OpenAI engineering deep dive into the Codex harness loop, prompt growth, tool-call replay, and stateless execution tradeoffs." picrew_harness,496,Catalog > Essential Readings & Ecosystem Maps,What We Learned Building Cloud Agents,https://cognition.ai/blog/what-we-learned-building-cloud-agents,,,,,,grey,"Cognition's field report on secure cloud-agent infrastructure, VM isolation, full-state snapshots, orchestration, governance, integrations, and enterprise adoption." picrew_harness,497,Catalog > Essential Readings & Ecosystem Maps,Writing effective tools for AI agents,https://www.anthropic.com/engineering/writing-tools-for-agents,,,,,,grey,Best practices for tool interface design so agents call tools safely and reliably. picrew_harness,498,Catalog > Essential Readings & Ecosystem Maps,"Your Agent Needs a Harness, Not a Framework",https://www.inngest.com/blog/your-agent-needs-a-harness-not-a-framework,,,,,,grey,Argument for reliability-first infrastructure around agents instead of framework-only thinking.