dmitryprg-ai/fix-last-task

Fix Last Task

Analyze failure before fixing. Patching without understanding why you missed the problem guarantees you'll miss it again.

When to Apply

• AI declared task complete, but user found errors
• Challenge protocol or Cross-check were not performed properly
• Completion report had high confidence but result is wrong

Workflow (7 Phases)

Phase 1: Analyze Failure

1. Restore original request (exact quote)
2. List what was done vs what was expected
3. Re-run Challenge protocol (Cross-check + 4 questions)
4. List all missed issues

Phase 2: Root Cause Analysis

Apply 5 Whys — dig at least 5 levels deep to find why issues were missed. Surface-level answers like "I forgot" are not root causes — keep asking until you reach a systemic reason.

Phase 3: Plan

Create TODO list — one item per missed issue, critical ones first.

Phase 4: Fix

For each TODO:

1. Mark in_progress
2. Apply fix
3. Cross-check: open file, read changes, verify
4. Mark completed
5. Report confidence change: [before]% -> [after]% — [reason]

Phase 5: Verify

Open and read EVERY changed file. Run the Challenge (4 questions) again.

Phase 6: Session Review

Record improvement to .cursor/data/improvements-backlog.md because patterns repeat without reflection — the same class of mistake will resurface unless codified.

Phase 7: Completion Report

Full DONE block: issues found, fixes applied, root cause, improvement recorded, final confidence.

What to Avoid

• Skipping phases — each exists because a specific class of error slipped through without it
• Fixing without Cross-check — you'll introduce new issues while fixing old ones
• Not recording improvement — the same mistake will repeat in 2-3 sessions
• Making excuses instead of analyzing — "I forgot" is not a root cause

Templates

For detailed analysis and review templates, see templates.md.