nmor/the-council/incident-investigation

Claude Code and agent skill by nmor/the-council

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Investigate service incidents and recurring customer failures by correlating logs, traces, recordings and durable state; distinguish mitigation from demonstrated root-cause fixes.

Install

npx vskill@latest install nmor/the-council/incident-investigation

SKILL.md

The same SKILL.md was published earlier as nmor/the-council/incident-investigation.

Incident investigation

Use for an outage, repeated customer failure or unexplained live regression. Reuse the existing incident record and plan. Keep containment and causal investigation distinct.

Build the evidence chain

Establish impact, time range, environment, deployed revision and affected/unaffected examples. Normalize timestamps with timezone and clock-skew caveats. Start with one representative failure and one comparison; expand only to test a hypothesis.

Follow a correlation ID across ingress, authentication, processing, dependencies, response delivery, durable writes and deferred work. Include both directions of a flow, retry attempts and finalization when applicable. Record missing spans rather than treating absence as proof that nothing happened. For voice incidents, compare audio, transcript and event timing; a transcript cannot establish silence, playout or interruption behavior. Redact customer data and credentials from shared artifacts.

Separate observed facts, hypotheses and ruled-out explanations. Specify a discriminating check before each investigation step. For latency, define caller-visible start/end boundaries and account for gaps outside component spans; do not add overlapping timings. For successful responses with missing effects, use data-reconciliation.

Fix and validate

Apply authorized containment with a measured exit condition. Reproduce the failure or explain what prevents reproduction. Test the smallest causal fix with a negative control, failure injection or representative replay; label simulated results as simulated. Recheck live impact when access permits. A disappearing alert or passing mock is insufficient to establish durable recovery. Use release-integration when multiple deployments must move together.

Deliverable and learning hooks

Record impact, causal timeline, evidence links, confidence, mitigation, permanent fix, verification and remaining unknowns in the existing record. Capture the detection or observability gap that would have shortened investigation; do not invent a root cause to close the incident.

References: Google SRE incident response and NIST SP 800-61 Rev. 3 for security incidents.

Size budget: 4 KB — token-budget.mjs --check.

Security

  • Security scanPassed, 95 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
12
Repository forks
2
7-day trend score
18

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/nmor/the-council/incident-investigation/badge)](https://verified-skill.com/skills/nmor/the-council/incident-investigation)

Something wrong with this skill or its page? Tell us.