# Monitorability Red Team

> Evaluate synthetic local multi-agent monitorability fixtures and label content, tool, state, and evidence blind spots

- Skill: `jmagly/monitorability-red-team` (Agent Skill)
- Install (CLI): `npx skillmds@latest add jmagly/monitorability-red-team`
- Raw SKILL.md: https://api.skillmd.com/api/skills/jmagly/monitorability-red-team/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: jmagly (https://skillmd.com/u/jmagly)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/jmagly/monitorability-red-team

---


# Monitorability Red Team

Run `aiwg monitorability-red-team run [fixture.json]` only on reviewed
synthetic fixtures. Require the fixture safety declaration, preserve captured
activity evidence, and preserve every `NOT RUN` reason.

Treat an undetected case without an observability label as a blocked integrity
failure. Do not infer production coverage from a passing synthetic report, and
do not introduce external inputs, provider calls, customer data, or live
channels into the fixture lane.

@implements #2045