quality-reflective-questions

Before marking ANYTHING as "done", ask yourself these questions and provide HONEST answers:

If you cannot answer ALL FOUR with specific, concrete details, the work is NOT complete.

❌ Bad (vague): "It's integrated" → ✅ Good (specific): "Imported in builder.py line 45" ❌ Bad (vague): "It works" → ✅ Good (specific): "Logs show execution at 10:30:45" ❌ Bad (vague): "Tests pass" → ✅ Good (specific): "46 unit tests + 2 integration tests pass"

Provides reflective questioning framework to challenge assumptions about work completeness, catching incomplete implementations before they're marked "done". Use before claiming features complete, before moving ADRs to completed status, during self-review, or when declaring work finished. Triggers on "is this really done", "self-review my work", "challenge my assumptions", "verify completeness", or proactively before marking tasks complete. Works with any type of implementation work. Enforces critical thinking about integration, testing, and execution proof. Source: dawiddutoit/custom-claude.