# h1-brain

An MCP server that gives an AI assistant a searchable local database of a HackerOne hunter's own bounty history plus 3,600+ public disclosed reports, and generates a full attack briefing for a chosen program.

h1-brain pulls a HackerOne user's own bug-bounty history, program scopes and report details from the HackerOne API into a local SQLite database, then exposes MCP tools that let a client such as Claude Desktop or Claude Code search, analyze and build on that history. It ships with a pre-built database of over 3,600 publicly disclosed, bounty-awarded reports drawn from the wider HackerOne community, including full write-ups, weakness types and bounty amounts.

Its primary tool, hack(handle), combines fresh program scope from the API with the user's past findings, public disclosures for that program, weakness patterns and untouched assets into one briefing formatted as actionable instructions for an AI assistant to use in an authorized bug-bounty engagement. Setup requires a Python virtual environment and a HackerOne API token.

Website: https://github.com/PatrikFehrenbach/h1-brain