MCP-Powered AI Research Engineer

Turn a single prompt into a full research report (web + your notes + code + charts).

](https://opensource.org/licenses/MIT) [ [](https://modelcontextprotocol.io/)

](https://github.com/prabureddy/ai-research-agent-mcp/pulls) [

[](https://github.com/prabureddy/ai-research-agent-mcp)

An autonomous AI agent that researches, codes, and writes comprehensive reports using the Model Context Protocol (MCP).

⚡ One‑liner install

Option 1: Using uv (Recommended - Fast!)

git clone https://github.com/prabureddy/ai-research-agent-mcp.git \
  && cd ai-research-agent-mcp/server \
  && uv venv \
  && source .venv/bin/activate \
  && uv pip install -r requirements.txt

Option 2: Using pip (Traditional)

git clone https://github.com/prabureddy/ai-research-agent-mcp.git \
  && cd ai-research-agent-mcp/server \
  && python3 -m venv venv \
  && source venv/bin/activate \
  && pip3 install -r requirements.txt

⚡ Quick Start: Clone → Install → Configure Claude Desktop → Start researching in 5 minutes!

📸 See It In Action

📺 [Watch the full video demo on YouTube](https://youtu.be/_cQLKpfDTBY)

Example Task → Output:

"Research top 3 programming languages in 2026 and create a comparison chart"

→ research_runs/2026-02-07_143022_programming-languages/
  ├── report.md              # Full analysis with sources
  ├── comparison_chart.png   # Visual comparison
  ├── data.json             # Raw statistics
  └── code/analysis.py      # Generated code

What the agent does: Searches web → Writes code → Creates charts → Generates report → Self-evaluates (all in ~60 seconds)

Who is this for?

• AI engineers doing research
• Indie hackers prototyping ideas
• Knowledge workers who live in Claude Desktop or Cursor

Overview

This system enables you to give a single command like:

"Compare electric scooters vs bikes for my commute and prototype a savings calculator"

And the agent autonomously:

• 🔍 Researches the web for relevant data
• 📚 Queries your personal knowledge base (notes, papers, docs)
• 💻 Writes and executes code (models, simulations, visualizations)
• 📊 Generates comprehensive reports with charts and analysis
• 🎯 Self-evaluates and logs quality metrics

Table of Contents

• Features
• Quick Start
• Architecture
• Configuration
• Usage
• Project Structure
• Troubleshooting
• Example Output
• Development
• Uninstallation
• Contributing
• License

Features