#99 · Primary category: Cybersecurity & Decryption Tools

OpenOSINT

ai-agent anthropic claude cli cybersecurity holehe llm maigret mcp mcp-server model-context-protocol osint osint-tools python reconnaissance security sherlock terminal threat-intelligence

AI-powered OSINT agent with interactive REPL, MCP server, and CLI. 19 tools. Works with Claude, GPT-4, or local models. For authorized security research only.

Project last updated:08/26/26

GitHub Stars

1.5K

Forks

238

Contributors

5

License

MIT

Why we included this project

Security researchers who would rather get answers than keep a dozen command-line tools straight will find OpenOSINT a useful middle layer. It wraps 19 investigation utilities, including email and username lookups, breach checks, WHOIS, subdomain enumeration, IP reputation, and dork generation, behind a natural-language interface you can run as a REPL, CLI, browser chat, or MCP server. The MCP option is what makes it interesting: any MCP-aware client, including Claude Desktop, can call the tools directly. The model picks which tool to invoke and chains the results, but the actual binaries run on your machine, so a hallucinated finding can't silently make it into the output. It works with Claude, GPT-4-compatible endpoints, or local Ollama models, and writes session reports as PDF or Markdown. One caveat: it's built for authorized investigations only, and several tools depend on third-party APIs and paid services.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category