Robin

Robin is an open-source, AI-powered OSINT tool for investigating the dark web. It searches Tor-accessible sources for a query, then uses a large language model to sift, summarize and organize what it finds, turning hours of manual digging into a structured investigation report. The project has about 7,600 stars on GitHub, is released under the MIT license, and was actively updated in 2026.

Overview

Dark web investigation is slow, manual work: onion services are hard to discover, pages are noisy, and most of what surfaces is irrelevant. Robin automates the tedious parts. Given a topic — a threat actor name, a leaked brand mention, a suspicious wallet — it queries multiple dark web search sources, fetches candidate pages through Tor, and passes the content to an LLM that filters noise and extracts the findings worth a human’s attention. The result is a readable summary of what is being said or sold in places that ordinary search engines never index. Robin is aimed at defensive research: threat intelligence, brand protection and early warning of leaks or planned attacks.

Key Features

• Automated searching of dark web sources through the Tor network
• LLM-powered filtering and summarization of noisy onion-site content
• Investigation-style output that highlights relevant findings and mentions
• Configurable queries for threat actors, brands, leaks and other topics
• Python-based and easy to run locally for sensitive research
• Open source under the MIT license
• Structured reporting that turns raw onion-site hits into an analyst-ready briefing
• Modular search backends let analysts add or swap dark web sources

💳 Pricing

Free. Robin is open source under the MIT license; you only provide your own LLM API access for the analysis step.

Who Is It For

Security researchers, threat-intelligence analysts, journalists and brand-protection teams who monitor dark web chatter and want AI help separating real signals from the noise.