Readerr - GitHub README Rating Agent

A production-ready AI agent that evaluates the quality of a GitHub repository's README and assigns a rating out of 5.0. It is built with Python, CrewAI, LangChain, Groq API (Llama 3.3 70B), and Rich for a beautiful CLI experience.

Features

• Automated Fetching: Automatically fetches the README content using the GitHub API without needing to clone the repo.
• AI Evaluation: Employs an expert AI Agent specializing in Open Source Documentation, Developer Experience, and Markdown Best Practices.
• Structured Scoring: Evaluates the README against 15 distinct criteria and computes a weighted overall rating.
• Actionable Feedback: Identifies strengths, weaknesses, missing sections, and top improvements.
• Beautiful CLI: Uses Rich to render an elegant terminal interface with tables, panels, and progress spinners.

Project Architecture

readme-rating-agent/
│── agents.py          # Defines the Evaluator Agent
│── tasks.py           # Defines the Evaluation Task and Pydantic Schema
│── tools.py           # Custom CrewAI tool to fetch README via GitHub API
│── run.py             # Main CLI entrypoint, Rich UI rendering, and orchestration
│── requirements.txt   # Project dependencies
│── .env.example       # Example environment variables
│── README.md          # Project documentation

Prerequisites

• Python 3.11+
• A valid Groq API Key for Llama 3 models.

Installation

1. Clone the repository (if you haven't already):

   git clone https://github.com/sarvan-2187/readme-rating-agent.git
   cd readme-rating-agent

2. Create a virtual environment (recommended):

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

   pip install -r requirements.txt

Configuration

1. Copy the example environment file:

   cp .env.example .env

2. Edit .env and add your Groq API key:

   GROQ_API_KEY=your_actual_api_key_here
   MODEL_NAME=llama3-70b-8192

Running the Application

Execute the CLI by running:

python run.py

You will be prompted to enter a GitHub repository URL:

Enter GitHub Repository URL: https://github.com/owner/repository

The agent will then begin processing the documentation. A spinner will indicate that the evaluation is in progress.

Example Output

Once finished, the CLI will display:

• Overall Rating: The final score out of 5.0 and star rating.
• Category Breakdown: A table with individual scores for Readability, Accessibility, Developer Experience, etc.
• Strengths & Weaknesses: A side-by-side comparison of the documentation's pros and cons.
• Actionable Feedback: Suggested missing sections and top improvements.
• Final Verdict: A comprehensive summary written by the AI Agent.

License

This project is licensed under the MIT License. Built by Sarvan Kumar