Summary

Robin is an AI-powered command-line tool designed for dark web Open Source Intelligence (OSINT) investigations. It assists users by leveraging Large Language Models (LLMs) to refine search queries, filter results from dark web search engines, and generate concise investigation summaries. Targeting security researchers, investigators, and automation enthusiasts, Robin aims to streamline the complex and often challenging process of gathering intelligence from the dark web.

How It Works

The tool employs a modular architecture, separating distinct workflows for searching, scraping, and LLM processing. It supports a variety of LLM backends, including OpenAI, Anthropic (Claude), Google (Gemini), and local models via Ollama, allowing users to select models based on their needs or available resources. Queries are refined by LLMs, and results are filtered to provide more relevant information, reducing manual effort and improving the efficiency of dark web investigations.

Quick Start & Requirements

• Prerequisites: Tor must be installed and running (e.g., apt install tor or brew install tor). API keys for supported LLM providers (OpenAI, Anthropic, Google) are required, configurable via a .env file or environment variables. For Ollama, a specific base URL configuration may be needed.
• Installation:
  • Docker (Recommended): docker run --rm -v "$(pwd)/.env:/app/.env" --add-host=host.docker.internal:host-gateway -p 8501:8501 apurvsg/robin:latest ui --ui-port 8501 --ui-host 0.0.0.0
  • Release Binary (CLI): Download the latest release binary, make it executable (chmod +x robin), and run via robin cli --model <model_name> --query "<your_query>".
  • Python (Development): Clone the repository, install dependencies (pip install -r requirements.txt), and run using python main.py -m <model_name> -q "<your_query>" -t <threads>.
• Links: No direct links to external documentation or demos are provided; the repository serves as the primary source.

Highlighted Details

• Modular Architecture: Clean separation of search, scrape, and LLM components for maintainability and extensibility.
• Multi-Model Support: Seamless integration with major LLM providers (OpenAI, Claude, Gemini) and local Ollama deployments.
• CLI-First Design: Optimized for terminal-based workflows and automation.
• Docker-Ready: Facilitates isolated and reproducible deployments.
• Custom Reporting: Option to save investigation summaries to files.
• Extensible: Designed for easy integration of new search engines, models, or output formats.

Maintenance & Community

The project welcomes contributions via pull requests and bug reports/feature requests via issues. Specific community channels (e.g., Discord, Slack) or details on active maintainers/sponsors are not detailed in the README.

Licensing & Compatibility

The README does not specify a software license. This absence makes it impossible to determine compatibility for commercial use or closed-source linking without further clarification.

Limitations & Caveats

The tool is intended strictly for educational and lawful investigative purposes, with a strong disclaimer against misuse and a reminder to comply with all relevant laws. Users must exercise caution when sending potentially sensitive queries through third-party LLM APIs. The functionality relies on the correct setup and background operation of the Tor network.