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.
npx skills add orlando-japan/claude-code-setting --skill runbook-authoringgit clone --depth 1 https://github.com/orlando-japan/claude-code-settingWrote 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.
[](https://agentmods.dev/skills/orlando-japan/claude-code-setting/runbook-authoring)<a href="https://agentmods.dev/skills/orlando-japan/claude-code-setting/runbook-authoring"><img src="https://agentmods.dev/badge/skills/orlando-japan/claude-code-setting/runbook-authoring/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.
<a href="https://agentmods.dev/skills/orlando-japan/claude-code-setting/runbook-authoring"><img src="https://agentmods.dev/badge/skills/orlando-japan/claude-code-setting/runbook-authoring.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00041 | $0.00938 |
| Opus 5 | $0.00020 | $0.00469 |
| Sonnet 5 | $0.00008 | $0.00188 |
| Haiku 4.5 | $0.00004 | $0.00094 |
Grade A, and why
runbook-authoring 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 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s https://metrics.internal/api/v1/query?query='rate(auth_errors[1m])' How it starts
The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Runbook authoring
A runbook is for tired people under pressure. Everything in it must be executable without context, without thinking, without making judgment calls the author forgot to remove.
Golden rule: the reader is you, at 3am, with a hangover, paged for the first time.
What a runbook is NOT
- A design doc.
- A description of how the system works.
- A narrative explaining the history of an incident.
- A document only the author can follow.
It's a checklist with copy-pasteable commands.
Template
Every runbook has the same sections:
1. When this page fires
The alert name that triggers this runbook, the condition, and the threshold. "You got paged because X is true."
2. What's broken
One sentence. Non-abstract. "The auth service can't reach Redis and is failing logins."
3. Immediate stabilization
The first thing to do before diagnosing. Rollback, kill switch, rate limit, failover. Commands that can be copy-pasted.
# Roll back the auth service
kubectl rollout undo deployment/auth -n prod
Not "roll back the deployment." The literal command.
4. Verify stabilization
How do you know the bleeding stopped? A specific metric, a specific query, a specific page to check.
# Error rate should drop below 1% in ~2 min
curl -s https://metrics.internal/api/v1/query?query='rate(auth_errors[1m])'
5. Diagnose
Now that things are stable (even if degraded), here's how to find what's really wrong. Dashboards to open (with links), logs to search (with exact queries), common causes to check.
This section is longer and can be a decision tree: "if X, then Y."
6. Fix
Once diagnosed, common fixes with exact commands.
7. Escalation
Who to wake up if you can't resolve it in N minutes. Names, rotations, contact methods.
8. After the page
- Where to file the incident report.
- Postmortem expectation.
- Links to related runbooks.
Rules
- Copy-pasteable commands. No
<your-service>placeholders. If the command needs a variable, provide a dropdown or a query to find it. - Dashboard and log links with the query pre-filled. Not "look at the dashboard." The literal URL with the time window.
- One decision at a time. No "this could mean several things, consider…" — walk the reader through the decision tree.
- Commands come with expected output. So the reader knows when it worked.
- Test it. Walk a teammate through it. If they ask questions, add the answers to the runbook.
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.
- 9d ago First seen · 102 lines · 41 tokens per session scan A 9fd5ecfd5b92
runbook-authoring is a skill published in the GitHub repository orlando-japan/claude-code-setting (2 stars, last pushed 3mo ago), licensed MIT. It adds 41 tokens to every session and 938 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.
Other skills, from other repositories
eng-runbook
A template for an engineering runbook, which is a practical guide engineers use during service operation and on-call incidents.
sre_triage
SRE first-response triage for distributed training incidents. Automates the manual checks from PyTorch/NCCL debugging runbooks.
runbook-authoring
Author operational runbooks so every alert maps to a tested, copy-pasteable, idempotent recovery procedure with diagnosis, remediation, escalation, rollback, and verification.
on-call-design
On-call rotation design, escalation tiers, runbook standards and burnout prevention.
building-incident-response-playbook
Designs and documents structured incident response playbooks that define step-by-step procedures for specific incident types aligned with NIST SP 800-61r3 and SANS PICERL frameworks. Covers playbook structure, decision trees, escalation criteria, RACI matrices, and integration with SOAR platforms. Activates for…
lastlight-debug
Debug a running Last Light instance over its admin API with the lastlight CLI — no SSH. Use when the user wants to "look at / debug a failed or stuck workflow", "why didn't the bot do X", "check a bad agent run / phase error on ", "tail a session", "search the logs", "read the server/docker logs", "list/trigger a…