plan-red-team

Attacks a technical plan, design, migration, rollout, or code change on paper before anything gets built or run, by interrogating its assumptions, surfacing failure modes and edge cases, rating every finding by severity, proposing safer alternatives, and stopping for the plan owner's explicit decision to proceed, revise, or cancel. Use when the user shares an architecture, refactor, rollout, vendor, data, or security decision and asks for a critique, risk review, pre-mortem, or sanity check before committing to it. Not for executing already-approved work, answering factual questions, chat with no technical system at stake, or post-incident postmortems.

Version

v1.0.0

Size

14 KB

Updated

2026-10-06

Your client verifies the signature before reading. This explicit manual read is available on every plan. Free automatically loads five chosen skills; paid plans automatically load relevant skills across the library.