Show HN: Red-team LLM reasoning and agent actions (honest scoring, local-first)

Show HN: Red‑team LLM reasoning and agent actions (honest scoring, local‑first) offers a ready‑to‑use toolkit for adding AI capabilities without building a mode