REA connects local reverse-engineering workflows to agents through MCP and a CLI. It analyzes authorized targets and returns evidence, provenance, coverage, and limitations for follow-up explanation or clean-room implementation.