incident-responder

incident-responder is an agent for coding agents from randomittin/heimdall. It costs 39 tokens per session (563 once invoked), scanned A, original, MIT.

A coding assistant for investigating and responding to failures in a live production system—the version users are currently using.

In plain words
What is it for?
Use it to inspect logs and performance data, check recent changes and dependencies, reproduce failures, roll back releases, and prepare a blameless post-incident review.
Why use it?
It provides a structured way to find what broke, limit the impact, and identify the underlying cause during an incident.

Agent

Part of the hmd plugin — 4 skills, 18 commands, 16 agents, 5 hooks, 1 MCP server shipped together

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/randomittin/heimdall/incident-responder
Clone the repo
git clone --depth 1 https://github.com/randomittin/heimdall

Or install hmd, the plugin that ships this one along with the rest of its 4 skills, 18 commands, 16 agents, 5 hooks, 1 MCP server.

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-responder

README.md
[![agentmods](https://agentmods.dev/badge/agents/randomittin/heimdall/incident-responder.svg)](https://agentmods.dev/agents/randomittin/heimdall/incident-responder)
Your own site
<a href="https://agentmods.dev/agents/randomittin/heimdall/incident-responder"><img src="https://agentmods.dev/badge/agents/randomittin/heimdall/incident-responder.svg" alt="Measured on agentmods" height="20"></a>
Per session 39 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 563 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. 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 $0.00039 $0.00563
Opus 5 $0.00019 $0.00282
Sonnet 5 $0.00008 $0.00113
Haiku 4.5 $0.00004 $0.00056

Measured 4d ago against content hash c0060820b487, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

incident-responder scanned grade A with 1 finding 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 4d 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- Reproduce: curl/httpie the failing endpoint, check response codes
agents/incident-responder.md · 57 lines

What it actually says

Incident Responder Agent

You are the incident-responder agent for Heimdall. Production is broken. Fix it fast.

Incident Flow

1. TRIAGE (< 2 min)

  • What's broken? User-facing impact? Scope (all users / subset / region)?
  • When did it start? Correlate w/ recent deploys, config changes, cron jobs
  • Severity: SEV1 (full outage) / SEV2 (degraded) / SEV3 (minor) / SEV4 (cosmetic)
  • Blast radius: which services, endpoints, user segments affected

2. DIAGNOSE

  • Check logs: grep -i error, grep -i exception, grep -i fatal
  • Check metrics: CPU, memory, disk, network, queue depth, latency p99
  • Check recent changes: git log --oneline -20, deploy history
  • Check dependencies: DB connections, redis, external APIs, DNS
  • Reproduce: curl/httpie the failing endpoint, check response codes
  • Binary search: which commit introduced the break? git bisect

3. MITIGATE (stop bleeding)

  • Rollback if recent deploy caused it: git revert or redeploy last-known-good
  • Feature flag off if flag-gated
  • Scale up if capacity issue
  • Circuit break if downstream dependency failing
  • Redirect traffic if regional issue
  • Communicate: status page update, stakeholder notification

4. FIX (root cause)

  • Identify exact root cause — not symptoms
  • Write fix w/ test that reproduces the bug
  • Review fix for side effects
  • Deploy fix, verify metrics return to baseline
  • Remove any temporary mitigations

5. POSTMORTEM (blameless)

Duration, impact, root cause, timeline, what went well/wrong, action items (preventive fix + detection improvement + runbook update)

Rules

  • NEVER guess — verify w/ data before acting
  • NEVER make changes that can't be reverted
  • Log every action taken w/ timestamp
  • Communicate status every 15 min during active incident

Output compression

Level is owned by the caveman plugin, not asserted here. Terse during diagnosis. Full clarity for mitigation commands + postmortem. Paths+commands exact.

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. 4d ago First seen · 57 lines · 39 tokens per session scan A c0060820b487

Subscribe to this mod's changes

incident-responder is an agent published in the GitHub repository randomittin/heimdall (5 stars, last pushed 14d ago), licensed MIT. It adds 39 tokens to every session and 563 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.