ops-workflows
Runbooks, blameless postmortems, change requests and on-call handoffs.
Install
/plugin marketplace add https://catylai.com/marketplace.json/plugin install ops-workflows@catylai
Two skills. runbook writes an operational runbook for a service or a
procedure, aimed at the person who will read it at 3am rather than at the person
who already knows how it works. incident-postmortem writes the blameless
write-up: timeline, root cause, contributing factors, what went well and where
you got lucky, and an action-item table.
Reach for it after an incident is resolved, or before one, when a service has no written procedure and the knowledge lives in one person's head.
It pairs with observability-core, which covers the live phase of an incident —
declaration, severity, blast radius, triage — and explicitly hands off to the
postmortem skill here at resolution. Use both: the core while the incident is
open, this one once it is closed.
In Claude Code the runbook skill searches your Terraform, Kubernetes and CI files for the facts it needs and writes the result to disk. On the web there is no checkout, so paste the details in instead.