Turn Claude Code into your SWE interview coach — behavioral, system design & coding training/mocks, all local.

A Claude Code plugin that reverse-engineers clean behavioral specs, test vectors, and acceptance criteria from any codebase, producing a provenance trail so a fresh team can reimplement without inheriting the original's internal structure.

Detects prompt injection by its effect on a sacrificial canary model, not just pattern matching: untrusted input hits a powerless model first, a behavioral check reads the residue, and it returns block, flag, or pass before your primary model acts. Inbound preflight sensor, not a guarantee.

Make every model cheaper or better. Zero-config behavioral harness for Claude Code, with a reproducible 408-run benchmark