Robin is an AI-powered dark web OSINT tool that combines language models with automated search and analysis workflows to investigate dark web content.
The tool addresses the challenge of conducting structured investigations on the dark web by automating query refinement and result analysis. It works by accepting user queries, leveraging large language models to optimize those queries for dark web search engines, executing searches through Tor, scraping results, and then using LLMs to filter findings and generate investigation summaries. The modular architecture separates search, scraping, and LLM processing into distinct workflows, allowing each component to function independently.
Robin suits investigators, security researchers, and analysts who need to conduct lawful dark web investigations with structured output and follow-up capabilities. The tool's Streamlit-based web interface makes it accessible without command-line expertise. Key distinguishing features include conversational follow-ups that answer questions grounded in a specific investigation's data without re-running searches, and one-click pivot suggestions that surface recommended follow-up queries from findings. The tool supports multiple LLM providers including OpenAI, Claude, Gemini, Ollama, and any OpenAI-compatible API, giving users flexibility in model selection. Docker deployment is recommended for isolated, clean usage. Anyone considering adoption should note the disclaimer that the tool is intended for educational and lawful purposes only, and users must comply with applicable laws and institutional policies before conducting investigations.
Maintainers respond to new issues and pull requests within a day. Work in the issue tracker centers on bug fixes, feature enhancements, and user questions.
