Document CI failure-signature triage runbook for JiraPS integration

[lipkau]

Why

A significant portion of integration recovery time was spent repeatedly mapping known failure signatures to likely causes.

Problem

Known Cloud/DC failure signatures are not documented in a single operator-focused runbook.

Impact

• Slower mean time to diagnosis for recurring failures
• Higher cognitive load for maintainers and future agents

Suggested approach

1. Add a concise CI triage runbook section that maps:
   • failure signature
   • likely root cause
   • first checks / next commands
2. Cover known patterns observed in recent runs (e.g., Jira DC JQL backend errors, transient version-delete 405 behavior).
3. Link from contributor/agent docs so this is used proactively.

Related areas

• AGENTS.md
• Tests/Integration/README.md
• CI workflow docs under .github/