| name | triage-issue |
| description | Analyze and triage issues — categorize, assess complexity, suggest approach. Use when: reviewing new issues, prioritizing backlog, or processing hub.issue relay events. |
triage-issue
Analyze and triage issues.
Trigger
- User asks to triage issues
bit issue list --open shows untriaged issues- Relay event:
hub.issue (optional, requires relay)
Steps
- Read the issue:
bit issue get <id> (skip session:<branch> labeled issues — they are session-coordination records, not backlog)
- Categorize: Determine type (bug, feature, question, improvement, docs)
- Assess complexity: Estimate as small/medium/large based on:
- Number of files likely affected
- Whether it requires new architecture
- Test coverage implications
- Suggest labels: Follow the label conventions in the bit-issue skill — one priority (
P0, P0.5, P1, P2, P3, P4) + area (merge, test, hub, perf, wasm, object, relay, sparse, …) + optional type (bug, edge-case, epic)
- Propose approach: Brief implementation plan if applicable
- Update issue: Add labels and comment with analysis
Output format
## Triage: <issue-title>
**Type**: bug | feature | question | improvement
**Complexity**: small | medium | large
**Labels**: P1, merge, bug
**Area**: modules/bit_core, modules/bit/cmd/bit, modules/bitx_hub, etc.
## Analysis
Brief description of what needs to be done.
## Suggested approach
1. Step 1
2. Step 2
3. Step 3
Commands
bit issue update <id> --label "P1" --label "merge" --label "bug"
bit issue comment add <id> --body "Triage: ..."
Important
- Don't close issues automatically
- For bugs, try to reproduce or identify the code path
- For features, check if similar functionality exists
- Link related issues if found
