incident-triage

incident-triage is a skill for Claude Code from skilletmd/skillet. It costs 47 tokens per session (639 once invoked), scanned A, original, Apache-2.0.

A structured guide for the first 15 minutes of responding to a production incident, such as an outage or serious service problem.

In plain words
What is it for?
Use it to assign an owner, set severity, check recent changes, roll back when appropriate, and begin diagnosis during an alert or outage.
Why use it?
It keeps the response focused on reducing user impact and communicating clearly before spending time looking for the underlying cause.

Skill for Claude Code

Written for Claude Code: user-invocable in frontmatter.

Good fit Use it to assign an owner, set severity, check recent changes, roll back when appropriate, and begin diagnosis during an alert or outage.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/skilletmd/skillet/incident-triage
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 skilletmd/skillet --skill incident-triage
Clone the repo
git clone --depth 1 https://github.com/skilletmd/skillet

Made for: Claude Code.

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-triage

README.md
[![agentmods](https://agentmods.dev/badge/skills/skilletmd/skillet/incident-triage/github.svg)](https://agentmods.dev/skills/skilletmd/skillet/incident-triage)
Your own site
<a href="https://agentmods.dev/skills/skilletmd/skillet/incident-triage"><img src="https://agentmods.dev/badge/skills/skilletmd/skillet/incident-triage/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-triage

Your own site · 80×15
<a href="https://agentmods.dev/skills/skilletmd/skillet/incident-triage"><img src="https://agentmods.dev/badge/skills/skilletmd/skillet/incident-triage.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 47 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 639 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 $0.00047 $0.00639
Opus 5 $0.00023 $0.00319
Sonnet 5 $0.00009 $0.00128
Haiku 4.5 $0.00005 $0.00064

Measured 11d ago against content hash cbb0f2a304c4, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

incident-triage 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 11d 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.

skills/incident-triage/SKILL.md · 55 lines

How it starts

The opening of the file, as written. The whole thing — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.

incident-triage

In an incident, the instinct is to find the root cause. That's the third job, not the first. Stop the bleeding, tell people, then diagnose. This skill keeps that order under pressure.

When to use

You're first to a production problem: an alert fired, a customer reported an outage, or something is visibly broken in prod.

Minute 0–2 — Declare and own it

  • Say it in the incident channel: "I'm on the API 500s, I have it." One owner, named, now.
  • Set a severity so everyone calibrates:
    • SEV1 — customer-facing outage or data at risk. Page others now.
    • SEV2 — degraded, some users affected. Fix this hour.
    • SEV3 — internal or cosmetic. File a ticket.

Minute 2–8 — Stabilize before you understand

You do not need the root cause to stop the bleeding.

  • Did it just start? Look at what changed: the last deploy, a feature flag, a config push, a traffic spike. Recent change is the cause far more often than not.
  • Can you roll back? A rollback you can do in 30 seconds beats a fix you understand in 30 minutes. Roll back first, diagnose after — but check for a forward-only database migration first; rolling code back over one can make things worse.
  • Can you shed load or turn the feature off? Flag off the broken path, rate-limit the hot endpoint, fail over to a replica.

Throughout — Communicate while you work

Post a one-line update every 10 minutes, even when there's no news:

[SEV1] API 500s on checkout. Rolled back deploy abc123 at 14:32.
Error rate dropping. Next update 14:45.

People fill silence with worst-case guesses. A boring update is worth ten Slack DMs you won't have time to answer.

Diagnose

Once it's stable, find the cause with evidence, not hunches:

  1. When did it start? Pin the timestamp from the metric, then find what shipped at that minute.
  2. What's the blast radius? One endpoint or all of them? One region, one customer, one tenant?
  3. Read the error, not the symptom. The 500 is the symptom; the stack trace or the saturated connection pool is the cause.

Read the full file on GitHub · 55 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. 11d ago First seen · 55 lines · 47 tokens per session scan A cbb0f2a304c4

Subscribe to this mod's changes

incident-triage is a skill published in the GitHub repository skilletmd/skillet (3 stars, last pushed 12d ago), licensed Apache-2.0. It adds 47 tokens to every session and 639 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.