chore: remove rag-8000.zcodecorp.in live links and hackathon text

fix: tighten reward/grade bounds to [0.01, 0.99] in models, orchestrator, and tests

fix: clamp scores to [0.01, 0.99] in API boundary, UI, inference, and tests

fix: strict score clamping to [0.01, 0.99] in all reward/grade paths

feat: fix image loading (LFS media URLs), improve Blog.MD tech depth, cleanup README

feat: add YouTube demo & GitHub repo links to README and Blog

Fix 400 error: extract detail from API errors, auto-reset on uninitialized env

Track large GIF with Git LFS

Add RAG OS demo GIF, fix image rendering (sanitize_html=False), de-dup blog wording

demo: upload agentic-rag-os.gif

Fix: use APP_ROOT env var for Docker path resolution (Blog.MD + image URLs)

Fix HF Space: add Blog.MD+assets to Docker, resolve image paths to GitHub raw URLs

Add Blog/Writeup and README tabs to Gradio UI for HF Space visibility

Revert "Add blog writeup sections to README for HF Space visibility: story, HF Space lab, RAG Master motivation, Agentic RAG OS origin, lessons learned"

Add blog writeup sections to README for HF Space visibility: story, HF Space lab, RAG Master motivation, Agentic RAG OS origin, lessons learned

Expand Blog.MD with HF Space story, RAG Master motivation, Agentic RAG OS origin; reorder links (RAG OS last) in README and Blog

Add Blog.MD, move plots to assets/qwen-finetuning-plots, update README badge & plot paths, update HF model card with training results

feat: 1GB storage limit, 2MB file limit, hash routing, landing animations

refactor: update ui design, remove github oauth support and ci.yml

fix: resolve frontend path, async retriever calls, and UI guards

chore: remove agentic-rag-os HF Space deployment

deploy: Agentic RAG OS live on HuggingFace Spaces

feat: add Agentic RAG OS — full-stack RL B2B RaaS platform

Revert "feat: add agentic_rag_os — RL B2B RaaS full-stack app - RAG pipeline (FAISS + sentence-transformers, per-project isolation) - Rewards-as-a-Service: 7 rule types, GRPO/PPO/DPO algorithms - GitHub OAuth 2.0 + JWT authentication - REST API: /auth, /users, /projects, /rewards (OpenAPI 3.0) - Full-stack UI: animated dark landing page + SPA dashboard - Export: JSON, CSV, HuggingFace dataset - Docker + GCP Cloud Run deployment configs (cloudbuild.yaml) - Architecture & deployment docs - Runs on port 8080, completely independent from gym server (:7860) - Updated README Wild Card section"

Revert "test: add HF Space gym API verification (52/52 passing)"

test: add HF Space gym API verification (52/52 passing)

feat: add agentic_rag_os — RL B2B RaaS full-stack app - RAG pipeline (FAISS + sentence-transformers, per-project isolation) - Rewards-as-a-Service: 7 rule types, GRPO/PPO/DPO algorithms - GitHub OAuth 2.0 + JWT authentication - REST API: /auth, /users, /projects, /rewards (OpenAPI 3.0) - Full-stack UI: animated dark landing page + SPA dashboard - Export: JSON, CSV, HuggingFace dataset - Docker + GCP Cloud Run deployment configs (cloudbuild.yaml) - Architecture & deployment docs - Runs on port 8080, completely independent from gym server (:7860) - Updated README Wild Card section

fix: isolate FAISS per domain, dynamic Auto Pilot step count

fix: Auto Pilot domain sync, Done flag, markdown status, README links

Merge pull request #1 from elizabeth07-m/main

docs: update readme.md - removed github link and MIT

feat: add legal_research domain, multi-domain switching, and README overhaul

chore: add git-lfs tracking for notebook and images

feat: GRPO fine-tuning pipeline + trained model for aerospace RAG

fix: mount Gradio at root, add Prometheus metrics

fix: delete dead database.py (sqlalchemy deps removed)

docs: update README baseline scores, add reproduce section, fix UI MySQL ref

test: add exploit tests, integration test, and CI workflow

fix: remove dead code (EXECUTOR, MySQL, bleach), fix PlannerAgent receiver

feat(domain): add legal research domain stub

fix(rewards+orchestrator): additive efficiency, repetition tracking, verify post-answer

fix(server): CORS, sanitization, concurrency, parameter override, mount at /ui

fix(graders): harden against keyword stuffing attacks

fix(openenv): add required image/port/entrypoint fields, add max_steps to tasks

fix(docker): remove hardcoded LLM vars, pre-warm embeddings, add .dockerignore

fix(inference): raise MAX_STEPS to per-task max, add answer forcing, fix fallback, save results

fix: add 30s LLM timeout + use HF Inference API for faster responses

fix: graceful LLM fallback + HF Inference API defaults in Docker

fix: Dockerfile copies README.md and source packages before pip install

fix: server tests with mock orchestrator injection - 89/89 passing