Skip to content

jchang98/deep_research_py

Repository files navigation

🐍 Deep Research Assistant PY

An AI-powered research tool in Python that helps you explore topics in depth using AI and web search.

Save 200 dollars a month and use this tool

⭐ A python port with a little more cli pizzazz of https://github.com/dzhng/deep-research

Contribute all you want to this. It was fun tweaking it.

video demo

alt text

Project Structure

deep_research_py/
├── run.py              # Main CLI interface
├── deep_research.py    # Core research logic
├── feedback.py         # Follow-up question generation
├── prompt.py           # System prompts for AI
└── ai/
    ├── providers.py    # AI service configuration
    └── text_splitter.py # Text processing utilities

Features

  • Interactive Research: Asks follow-up questions to better understand your needs
  • Depth Control: Customize research breadth and depth
  • Web Integration: Uses Firecrawl for reliable web content extraction
  • Smart Synthesis: Combines multiple sources into coherent findings
  • Beautiful CLI: Rich text interface with progress tracking
  • Markdown Reports: Generates well-formatted research reports

Installation

uv tool install deep-research-py && cp .env.example .env

Configuration

Open .env and replace placeholder values with your actual API keys

Set up environment variables in .env file:

# Required by service: "deepseek" and "openai"
# unless you're using DeepSeek or another OpenAI-compliant API.
OPENAI_API_KEY=your-openai-key-here

# Optional: ollama related environment variable
# OLLAMA_API_ENDPOINT=http://localhost:11434

# Required: Firecrawl API key
FIRECRAWL_API_KEY=your-firecrawl-key-here
# If you want to use your self-hosted Firecrawl, add the following below:
# FIRECRAWL_BASE_URL="http://localhost:3002"

Note: If you prefer, you can use DeepSeek instead of OpenAI. You can configure it in the .env file by setting the relevant API keys and model. Additionally, ensure that you set DEFAULT_SERVICE to "deepseek" if using DeepSeek or "openai" if using OpenAI.

Note: If you want use ollama as your LLM service, set DEFAULT_SERVICE to "ollama" or use --service parameter.

Usage

Run the research assistant:

deep-research

You'll be prompted to:

  1. Enter your research topic
  2. Set research breadth (2-10, default 4)
  3. Set research depth (1-5, default 2)
  4. Answer follow-up questions
  5. Wait while it researches and generates a report

You can change the concurrency level by setting the --concurrency flag (useful if you have a high API rate limit):

deep-research --concurrency 10

You can get a list of available commands:

deep-research --help

Development Setup

Clone the repository and set up your environment:

# Clone the repository
git clone https://github.com/epuerta9/deep-research-py.git
cd deep-research-py

# Create and activate virtual environment
uv venv
source .venv/bin/activate

# Install in development mode
uv pip install -e .

# Copy environment configuration
cp .env.example .env

# Set your API keys by editing the .env file

# Run the tool
deep-research

Requirements

  • Python 3.9 or higher
  • OpenAI API key (GPT-4 access recommended)
  • Firecrawl API key for web search
  • Dependencies:
    • openai
    • firecrawl-py
    • typer
    • rich
    • prompt-toolkit
    • aiohttp
    • aiofiles
    • tiktoken

Output

The tool generates:

  • A markdown report saved as output.md
  • List of sources used
  • Summary of key findings
  • Detailed analysis of the topic

License

MIT

Contributing

Contributing

Contributions are welcome! Please follow these steps:

  1. Fork the repository
  2. Create your feature branch (git checkout -b feature/amazing-feature)
  3. Install development dependencies:
pip install pre-commit
pre-commit install
  1. Make your changes
  2. Commit your changes (git commit -m 'Add amazing feature')
  3. Push to the branch (git push origin feature/amazing-feature)
  4. Open a Pull Request

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages