Triage

Workflow: Triage

Assess a new issue and prepare it for the backlog.

When to Use

• New issue arrives without full context, priority, or layer identification
• Issue needs scoping before it can be picked up by an engineer

Inputs

• Issue identifier, title, and description
• Any linked parent issues or related tickets
• The target repo for context

Steps

1. Read the issue — understand what’s being requested
2. Classify — determine issue type (feature, bug, tech debt, infra)
3. Identify delivery layers — which of backend, frontend, api, infra, docs are affected
4. Assess priority — critical / high / medium / low based on impact and urgency
5. Check for duplicates — search existing issues for overlap
6. Enrich the issue:
   • Add missing delivery layers
   • Add priority if not set
   • Add acceptance criteria if missing
   • Link to relevant PRDs, designs, or ADRs
7. Comment with triage summary: type, layers, priority, and any questions for the reporter

Quality Checks

• Delivery layers are identified (required before moving to todo)
• Priority is set
• Description answers: what, why, and what does done look like
• No duplicate of an existing issue

References

• See “When a Ticket Goes Into Backlog” and “When a Ticket Is Ready to Build” in .ai/projects/cascadeguard/sdlc.md