Spaces:
Sleeping
Sleeping
File size: 3,327 Bytes
9677321 6c22dda f09fb58 9677321 d72844a | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 | ---
title: Meta_com OpenEnv Agent
emoji: 🤖
colorFrom: blue
colorTo: indigo
sdk: docker
app_port: 7860
app_file: server/app.py
pinned: false
---
<div align="center">
<img src="https://img.shields.io/badge/Status-Beta-blue?style=for-the-badge" alt="Status Badge">
<img src="https://img.shields.io/badge/Platform-Hugging_Face-orange?style=for-the-badge" alt="HF Badge">
<img src="https://img.shields.io/badge/OpenEnv-Compatible-brightgreen?style=for-the-badge" alt="OpenEnv Badge">
</div>
# Meta_com: Autonomous Git Conflict Resolution Agent
> An advanced, OpenEnv-compliant reinforcement learning and large language model benchmark environment for autonomous Git Merge Conflict resolution.
Welcome to **Meta_com**. This project transforms one of software engineering's most manual struggles into a reproducible, programmable environment for evaluating AI Agents.
---
## 🌟 Key Features
- **Pioneering OpenEnv Structure:** Out-of-the-box readiness with the standard `[START]`, `[STEP]`, and `[END]` evaluation logs.
- **FastAPI Backend:** Built-in web server with robust validation capabilities (`POST /reset`) natively tested on Hugging Face Spaces.
- **Gradio User Interface:** Sleek, interactive visualization dashboard to run test evaluations right from the browser.
- **Modular Reinforcement Engine:** Provides dense rewards via semantic evaluations, unit tests, and syntax validators.
---
## 🛠️ Architecture and Stack
The deployment seamlessly bridges Python AI Agents and public accessibility components:
1. `app.py`: High-performance FastAPI server exposing validation APIs, mounting the interactive Gradio UI.
2. `inference.py`: Core AI execution script communicating with OpenAI-compatible endpoints to resolve complex OpenEnv inputs.
3. `my_env_v4.py`: The isolated git-sandbox environment ensuring deterministic states across validations.
---
## 🚀 How to Validate and Run Tests
To ensure the agent meets the strictest standards, a professional validation script is included.
### Testing Locally
Run the visual interface and API on your machine:
```bash
# 1. Install Dependencies
pip install -r requirements.txt
# 2. Launch Local Space
uvicorn app:app --host 0.0.0.0 --port 7860
```
Then visit `http://localhost:7860` to access the full Inference UI.
### Automated Cloud Validation
If you've pushed this branch to a Hugging Face Space configured with our `Dockerfile`, execute the tester locally against your cloud deployment:
```bash
chmod +x scripts/validate-submission.sh
./scripts/validate-submission.sh https://your-username-your-space.hf.space
```
**Validation Checks Ensure:**
- ✅ Valid `HTTP POST /reset` communication bindings exist.
- ✅ Remote Dockerization succeeds without error boundaries.
- ✅ Standard `openenv validate` protocols pass fully.
---
## 📚 Technical Implementation Phases
| Level | Description | Complexity |
|---|---|---|
| **Phase 1** | Trivial merges. Agents simply ingest blocks from both code sources. | Easy |
| **Phase 2** | Multi-file dependency evaluation. Requires cross-file API signature knowledge. | Medium |
| **Phase 3** | Deep logic integration. Competing intent behaviors mapped and synthesized. | Hard |
*Developed exclusively under rapid iteration cycles. Powered by Python 3.10+, FastAPI, and OpenEnv Core Frameworks.*
|