GitHub Repository Summarizer API

A FastAPI service that analyzes GitHub repositories and generates intelligent summaries using Large Language Models (LLMs).

Features

• Analyzes public GitHub repositories
• Web UI at http://localhost:8000 — no curl needed
• Detailed summaries: overview, technologies, architecture, key features, use cases, and getting started guide
• Intelligent file filtering and prioritization
• Context-aware LLM prompting
• Support for multiple LLM providers (Nebius, Anthropic, OpenAI)
• GitHub token support to avoid API rate limits
• Comprehensive error handling
• RESTful API with proper HTTP status codes

Setup Instructions

Prerequisites

• Python 3.10 or higher
• pip package manager
• Git (optional, for cloning)

Installation

1. Download and extract the project (or clone if using git):

# If cloning:
git clone <your-repo-url>
cd github-summarizer

# If from zip:
unzip github-summarizer.zip
cd github-summarizer

2. Create a virtual environment (recommended):

python -m venv venv

# On macOS/Linux:
source venv/bin/activate

# On Windows:
venv\Scripts\activate

3. Install dependencies:

pip install -r requirements.txt

4. Configure API keys:

Create a .env file in the project root:

# Copy the example file
cp .env.example .env

Then edit .env and add your keys:

# LLM API key (required) — pick one:
NEBIUS_API_KEY=your-actual-nebius-api-key-here
# ANTHROPIC_API_KEY=your-actual-anthropic-api-key-here
# OPENAI_API_KEY=your-actual-openai-api-key-here

# GitHub token (recommended) — avoids GitHub API rate limits
# Generate at: https://github.com/settings/tokens
GITHUB_TOKEN=your-github-token-here

Note: Without a GITHUB_TOKEN you are limited to 60 GitHub API requests per hour. With a token the limit rises to 5,000.

Note: The .env file is automatically ignored by git (see .gitignore) to keep your keys secure.

Running the Server

source venv/bin/activate
uvicorn main:app --host 0.0.0.0 --port 8000

The server starts at http://localhost:8000. You should see:

INFO:     Started server process
INFO:     Waiting for application startup.
INFO:     Application startup complete.
INFO:     Uvicorn running on http://0.0.0.0:8000

Press Ctrl+C to stop it.

Usage

Web UI (recommended)

Open your browser and go to:

http://localhost:8000

Paste any public GitHub URL into the input box and click Summarize. Results appear on the page.

API (curl)

Endpoint: POST /summarize

curl -s -X POST http://localhost:8000/summarize \
  -H "Content-Type: application/json" \
  -d '{"github_url": "https://github.com/psf/requests"}' | python3 -m json.tool

Response (example):

{
  "summary": "...",
  "technologies": ["Python", "urllib3", "..."],
  "structure": "...",
  "key_features": ["...", "..."],
  "architecture": "...",
  "getting_started": "...",
  "use_cases": ["...", "..."]
}

Error Response:

{
  "detail": "Repository not found or is private"
}

Health Check

curl http://localhost:8000/health

Interactive API Docs

FastAPI provides a built-in UI for exploring the API:

http://localhost:8000/docs

Model Selection

Chosen Model: Qwen/Qwen3-Coder-30B-A3B-Instruct (Nebius AI Studio)

Why this model?

• Specialized for code: Qwen3-Coder is specifically trained for code understanding and analysis
• Strong reasoning: Provides excellent analysis capabilities for diverse codebases
• Cost-effective: Available through Nebius AI Studio
• Fast: Optimized for quick responses while maintaining quality

The implementation also supports Anthropic Claude and OpenAI GPT-4 as alternatives.

Repository Processing Strategy

What We Include

1. Priority Files (always included):

   • README files (README.md, README.rst, etc.)
   • Package configuration (package.json, setup.py, pyproject.toml, Cargo.toml, etc.)
   • Build files (Dockerfile, Makefile, docker-compose.yml)
   • Architecture documentation (ARCHITECTURE.md, DESIGN.md)

2. Source Files (selectively):

   • Root-level source files (up to 10 files)
   • Main source files from primary directories (up to 15 files)
   • Prioritizes common languages: .py, .js, .ts, .java, .go, .rs, .c, .cpp

3. Configuration Files (limited):

   • Key config files like tsconfig.json, .eslintrc (up to 5 files)

4. Directory Structure:

   • Complete file tree (filtered, limited to 100 items)

What We Skip

1. Binary Files: Images, videos, audio, compiled binaries, archives, fonts
2. Generated/Dependency Files: node_modules/, __pycache__/, lock files, build artifacts
3. Environment/IDE Files: .vscode/, .idea/, virtual environments

Context Management

• Individual file limit: 5,000 characters (with truncation marker)
• Total context limit: ~100,000 characters across all files
• Smart selection: Priority files processed first
• Graceful degradation: System continues even if some files fail to fetch

Project Structure

.
├── main.py              # FastAPI application, endpoints, and web UI
├── repo_analyzer.py     # GitHub API integration and file filtering
├── llm_client.py        # LLM API integration (multi-provider)
├── test_api.py          # API test suite
├── requirements.txt     # Python dependencies
└── README.md            # This file

Error Handling

HTTP Status	Meaning
400 Bad Request	Invalid GitHub URL format
404 Not Found	Repository doesn't exist or is private
500 Internal Server Error	API key not configured or unexpected error
503 Service Unavailable	Network error or LLM API unavailable

Troubleshooting

LLM API key not configured

• Make sure .env exists and contains NEBIUS_API_KEY, ANTHROPIC_API_KEY, or OPENAI_API_KEY

rate limit exceeded

• Add a GITHUB_TOKEN to your .env file (see Setup)

Repository not found or is private

• Verify the URL is correct and the repository is public

Connection timeout

• Check your internet connection and try again

Module not found errors

• Make sure you activated your virtual environment: source venv/bin/activate

License

This project is provided as-is for evaluation purposes.