ContextMesh 🕸️

Living Documentation & Code Analysis Powered by Gemini 1.5 Pro

ContextMesh is a cutting-edge "Lean Monolith" application designed to bridge the gap between code and documentation. Leveraging the massive 2M token context window of Google's Gemini 1.5 Pro, it ingests entire repositories to provide holistic analysis, refactoring advice, and automated documentation.

---

The Agents

ContextMesh exposes three distinct AI personas to interact with your codebase:

The Architect

Analyzes high-level patterns, dependencies, and architectural health.

• Identifies circular imports and scalability bottlenecks.
• Generates "Health Scores" for Modularity, Documentation, and Test Coverage.

The Refactorer

Your dedicated technical debt collector.

• Identifies messy code and anti-patterns.
• Provides specific "Before" and "After" refactoring plans.
• Predicts the impact of proposed changes.

The Documentarian

Turns code into plain English.

• Generates comprehensive READMEs and "Living Documentation".
• Explains data flow and key architectural decisions, not just syntax.

---

Tech Stack

Backend

• Framework: FastAPI (Python)
• AI Model: Google Gemini 1.5 Pro (via google-genai)
• Utilities: pathspec (gitignore handling), pydantic (validation)
• Integration: GitHub Webhooks

Frontend

• Framework: Next.js 16 (App Router)
• Library: React 19
• Styling: Tailwind CSS v4
• Icons: Lucide React

---

Getting Started

Prerequisites

• Python 3.10+
• Node.js 18+
• A Google Cloud Project with Gemini API enabled (Get API Key)

Installation

1. Clone the Repository

   git clone https://github.com/yourusername/contextmesh.git
   cd contextmesh

2. Backend Setup

   # It is recommended to use a virtual environment
   python -m venv venv
   source venv/bin/activate  # Windows: venv\Scripts\activate
   
   # Install Python dependencies
   pip install -r backend/requirements.txt
   
   # Set your API Key
   # Linux/Mac
   export GEMINI_API_KEY="your_actual_api_key_here"
   # Windows (PowerShell)
   $env:GEMINI_API_KEY="your_actual_api_key_here"

   (Alternatively, create a .env file in the root directory)

3. Frontend Setup

   cd frontend
   npm install

---

Running the Application

You will need two terminal windows to run the full stack.

1. Start the Backend API

From the project root:

uvicorn backend.main:app --reload --port 8000

• API URL: http://localhost:8000
• Interactive Docs: http://localhost:8000/docs

2. Start the Frontend Dashboard

From the frontend/ directory:

npm run dev

• Dashboard: http://localhost:3000

---

API Endpoints

• POST /analyze: Main endpoint to trigger an agent (Architect, Refactorer, Documentarian) on a local path.
• POST /webhook: Webhook receiver for GitHub events.
• GET /events: Fetch recent webhook events.