incident-response

incident-response is a skill for Claude Code, Codex from orlando-japan/claude-code-setting. It costs 36 tokens per session (853 once invoked), scanned A, original, MIT.

A step-by-step guide for handling incidents in a live production system, such as outages or user-facing regressions. It covers stopping the damage, finding the cause, fixing it, and learning from it.

In plain words
What is it for?
Use it during outages, after being paged, or when users report a regression. It guides rollbacks, feature-flag changes, traffic controls, communication, investigation, and follow-up.
Why use it?
It gives responders a clear order of operations when a live system is failing. This helps avoid spending early response time diagnosing while the problem continues to spread.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it during outages, after being paged, or when users report a regression. It guides rollbacks, feature-flag changes, traffic controls, communication, investigation, and follow-up.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/orlando-japan/claude-code-setting/incident-response
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add orlando-japan/claude-code-setting --skill incident-response
Clone the repo
git clone --depth 1 https://github.com/orlando-japan/claude-code-setting

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for incident-response

README.md
[![agentmods](https://agentmods.dev/badge/skills/orlando-japan/claude-code-setting/incident-response/github.svg)](https://agentmods.dev/skills/orlando-japan/claude-code-setting/incident-response)
Your own site
<a href="https://agentmods.dev/skills/orlando-japan/claude-code-setting/incident-response"><img src="https://agentmods.dev/badge/skills/orlando-japan/claude-code-setting/incident-response/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for incident-response

Your own site · 80×15
<a href="https://agentmods.dev/skills/orlando-japan/claude-code-setting/incident-response"><img src="https://agentmods.dev/badge/skills/orlando-japan/claude-code-setting/incident-response.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 36 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 853 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00036 $0.00853
Opus 5 $0.00018 $0.00426
Sonnet 5 $0.00007 $0.00171
Haiku 4.5 $0.00004 $0.00085

Measured 9d ago against content hash 433c606ee95d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

incident-response scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

templates/extra/skills/incident-response/SKILL.md · 63 lines

How it starts

The opening of the file, as written. The whole thing — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Incident response

Order matters: stop the bleeding first, diagnose second, fix third, learn fourth. Skipping steps costs minutes you don't have.

Phase 1 — Stabilize (minutes 0–15)

Goal: stop the damage growing. You do not need to understand why yet.

  • Roll back if the incident began after a deploy. Rollback beats diagnosis in the first 15 minutes. Always.
  • Disable the broken path. Feature flag off. Route traffic away. Kill the bad worker.
  • Rate-limit upstream if the system is drowning.
  • Announce. Post to the incident channel: what you see, what you're doing, ETA to next update.
  • Appoint a single driver if more than one person joins. Confusion multiplies.

Phase 2 — Diagnose (minutes 15–60)

Once stable (even if degraded), find the cause.

  • Check dashboards first. Which metric turned red, when, and how? That's your start-time and scope.
  • Check logs at the transition. Not now-logs, transition-logs. What's different at the moment things broke?
  • Correlate with deploys, feature flag changes, config pushes, infra events. Incidents correlate with change.
  • Test the hypothesis before claiming you know. "It's the DB" means nothing without evidence.

Phase 3 — Fix (minutes 60+)

  • Patch, don't rewrite. Smallest possible fix. "Add a null check" beats "refactor the auth module."
  • Verify the fix locally or in staging before deploying, if you can without escalating the incident.
  • Deploy carefully. Partial rollout, watch dashboards. Don't fix on deploy auto-pilot.
  • Confirm recovery. The metric you measured in phase 2 should return to baseline.

Phase 4 — Learn

After the system is stable. Not the same day necessarily, but within a week.

  • Timeline. When did each event happen? Use actual timestamps, not "around noon."
  • What went well. Rollback worked. Alerts fired on time. Someone caught the mistake early.
  • What went badly. Blind spots. Slow response. Misleading dashboards.
  • Root causes. 5 Whys. Don't stop at the technical cause ("a null pointer") — go to the process cause ("our tests didn't cover this path because it's only triggered by data we never see in staging").
  • Action items. Concrete, owned, with a deadline. "Improve monitoring" is not an action item. "Alert on p99 > 500ms for /api/auth by Friday, owner: @Alice" is.

Read the full file on GitHub · 63 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 9d ago First seen · 63 lines · 36 tokens per session scan A 433c606ee95d

Subscribe to this mod's changes

incident-response is a skill published in the GitHub repository orlando-japan/claude-code-setting (2 stars, last pushed 3mo ago), licensed MIT. It adds 36 tokens to every session and 853 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.