| name | write-ops-log |
| description | Publish a tagged postmortem incident record to Echo. Use after an infrastructure or durable debugging session, then link the canonical Echo URL from the associated PR or issue. |
Skill: Write an Ops Log
Publish the incident record to Echo. Do not add a repository debug-log file.
The audience is a future engineer who must quickly decide what to check or do
next. Record the smallest set of facts that supports that action. Link raw
evidence and detailed chronology rather than reproducing them; Echo entries
consume limited model context whenever they are retrieved.
Search before writing
Invoke consult-echo and run its complete search-before-write sequence. Edit
the existing entry when it covers the same incident. Start with a natural-language
infra/echo/cli.py search, fetch likely matches with get <domain:id>, and use
grep for the exact error or run identifier. Create a new entry for a different
incident even when the symptom resembles an older one; link related incidents
and create or extend a separate synthesis only when they establish a reusable
cross-incident pattern.
Draft the Echo entry
Write an OKF document in a temporary file:
---
type: wiki-note
title: "Incident YYYY-MM-DD: <system> — <symptom>"
use_when: when investigating <specific symptom or exact error>
tags:
- incident
- debugging
- ops
- <system>
-
-
---
# :
## TL;DR
-
-
-
## Response
## Cause and resolution
## Artifacts
-
