Coding Research AI Agent

Discover, Compare, and Analyze Developer Tools with AI

Overview

The Coding Research AI Agent is an intelligent CLI tool that helps developers discover and compare alternative tools, libraries, and platforms. Using advanced AI workflows, it automatically researches tools, scrapes documentation, and provides personalized recommendations based on your specific needs.

Quick Start

Prerequisites

• Python 3.13+
• uv package manager

Installation

1. Clone the repository

   git clone https://github.com/yourusername/coding-research-agent.git
   cd coding-research-agent

2. Install dependencies with uv

   uv sync

3. Run the application

   uv run main.py

Configuration

Create a .env file in the project root:

# Required API Keys
ANTHROPIC_API_KEY=your_anthropic_api_key_here
FIRECRAWL_API_KEY=your_firecrawl_api_key_here

Getting API Keys

• Anthropic API Key: Sign up at console.anthropic.com
• Firecrawl API Key: Get yours at firecrawl.dev

Usage Examples

Basic Queries

Question: React alternatives
Question: databases better than MySQL
Question: hosting platforms like AWS

Specific Use Cases

Question: vector databases for AI applications
Question: free alternatives to GitHub Copilot
Question: Python web frameworks faster than Django

Sample Output

Project Structure

coding-research-agent/
├── app/
│   ├── __init__.py
│   ├── workflow.py          # Main LangGraph workflow orchestration
│   ├── models.py            # Pydantic data models
│   ├── prompts.py           # AI prompts and dynamic categorization
│   ├── logger.py            # Progress logging and CLI interface
│   └── firecrawl.py         # Web scraping service with retry logic
├── media/                   # README media files
├── main.py                  # CLI entry point
├── pyproject.toml          # uv project configuration
├── .gitignore              # Git ignore rules
└── README.md               # This file

Development

Running with uv

# Activate virtual environment
uv venv

# Install new dependencies
uv sync

# Run the application
uv run main.py

Available Commands

• help - Show help menu with examples
• clear - Clear the terminal screen
• exit/quit - Exit the application

Technical Architecture

Core Components

1. LangGraph Workflow - Orchestrates the research pipeline with three main nodes:

   • extract_tools - Finds and extracts tool names from articles
   • research - Gathers detailed information about each tool
   • analyze - Generates personalized recommendations

2. Dynamic AI Categorization - Uses Claude to automatically:

   • Detect query categories
   • Generate relevant examples for better extraction
   • Exclude generic terms to improve accuracy

3. Fallback System - Multiple strategies ensure reliable results:

   • Article-based extraction
   • Direct search results
   • AI-generated suggestions