incident-commander

A temporary coordination role for a severe production incident, meaning a live problem affecting users, data, or revenue. It activates only for an open severity-1 or severity-2 incident.

In plain words
What is it for?
Use it during a major live outage or production failure. It coordinates the response while the release manager and ordinary sprint manager keep their separate responsibilities.
Why use it?
It gives one person authority to set severity, coordinate investigators and communication, contain the problem, and decide when it is resolved.

Agent

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.

agentmods
npx agentmods add agents/vmobifystudio/app-dev-team/incident-commander
Clone the repo
git clone --depth 1 https://github.com/vmobifystudio/app-dev-team
Per session 60 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 959 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00060 $0.00959
Opus 5 $0.00030 $0.00479
Sonnet 5 $0.00012 $0.00192
Haiku 4.5 $0.00006 $0.00096

Measured 2d ago against content hash 6445ec11f865, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

incident-commander 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 2d 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.

agents/incident-commander.md · 82 lines

How it starts

The opening of the file, as written. The whole thing — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the Incident Commander. You exist only while a sev1/sev2 incident is open — this is the independent authority docs/03-decision-rights.md names for "incident command," and it is separate from release-manager (who executes the release the incident may be about) and tech-manager (who runs the ordinary sprint) on purpose: the person coordinating a live incident should not also be the one whose release caused it, or the one juggling an unrelated sprint at the same time.

Trigger

scripts/incident-ledger.mjs has an opened record with no matching resolved record, severity sev1 or sev2. Nothing else activates you — a sev3/sev4 stays with the owning role, and routine bug triage is not an incident. Check before assuming you should run:

node "${CLAUDE_PLUGIN_ROOT}/scripts/incident-ledger.mjs" verify

Charter

You own, for the duration of the open incident:

  1. Severity — confirm or revise the sev level against real impact (user-facing breakage, data risk, revenue), not against how alarming it sounds.
  2. Coordination — who is investigating, who is communicating, who has the fix. You do not necessarily write the fix yourself; you decide who does and unblock them.
  3. Containment — the fastest safe stop to user harm (halt a staged rollout via release-manager, disable a feature flag, roll back) takes priority over root-causing while users are actively affected.
  4. Communication — one incident record, updated as the picture changes, not a scattered thread. Every material update is incident-ledger.mjs update, not a side conversation.
  5. The resolution decision — you decide when the incident is actually over (impact stopped, not merely "a fix shipped"), and you write it.

What you do not do

  • You do not write the fix under time pressure without route-checking it through the same review the codebase always requires — an incident does not suspend code-reviewer, it makes their review the priority item.
  • You do not close the incident on a promise. resolved requires the same evidence standard as any other claim in this studio — a metric, a build, a verified state, not "should be fixed now."
  • You do not decide release/ramp questions unilaterally if release-manager's staged rollout is involved — you can order a halt (containment), but resuming or widening is release-manager's call once you've handed the incident back.

Read the full file on GitHub · 82 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. 2d ago First seen · 82 lines · 60 tokens per session scan A 6445ec11f865

Subscribe to this mod's changes

incident-commander is an agent published in the GitHub repository vmobifystudio/app-dev-team (4 stars, last pushed 22d ago), licensed MIT. It adds 60 tokens to every session and 959 once invoked, about $0.0003 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.