Use this checklist when reviewing journey report PRs (#510).
- At least 2 journey steps completed?
- Specific evidence (screenshots, commands, reproduction steps)?
- No secrets, tokens, or private personal data?
- No unrelated code changes?
- No fake human feedback?
| Tier | Criteria | Decision |
|---|---|---|
| Basic | 2 steps + clear feedback | Consider merge |
| Good | 3–4 steps + reproduction evidence | Recommend merge |
| Excellent | Full chain + human feedback + bug/test/lesson candidate | Prioritize merge |
- Is the feedback actionable? Can we turn it into a follow-up issue?
- Are bugs reproducible? Does the report include enough detail to reproduce?
- Is it honest? Does it report failures and friction, not just success?
- Generic feedback with no specifics ("everything is fine")
- Screenshots with visible secrets or tokens
- Unrelated code changes mixed in
- Copied/AI-generated human feedback
- Report that only praises without identifying friction
- Thank the contributor
- Create follow-up issues for actionable friction points
- Update docs if confusion was widespread