---

name: verification-before-completion description: | Use when about to claim work is complete, fixed, or passing. REQUIRES running verification commands and confirming output before making any success claims. Evidence before assertions, always.

Verification Before Completion

Overview

Claiming work is complete without verification is dishonesty, not efficiency.

Core principle: Evidence before claims, always.

Violating the letter of this rule is violating the spirit of this rule.

The Iron Law

CopyNO COMPLETION CLAIMS WITHOUT FRESH VERIFICATION EVIDENCE

If you haven't run the verification command in this message, you cannot claim it passes.

The Gate Function

CopyBEFORE claiming any status or expressing satisfaction:

1. IDENTIFY: What command proves this claim?
2. RUN: Execute the FULL command (fresh, complete)
3. READ: Full output, check exit code, count failures
4. VERIFY: Does output confirm the claim?
   - If NO: State actual status with evidence
   - If YES: State claim WITH evidence
5. ONLY THEN: Make the claim

Skip any step = lying, not verifying

Common Failures

Red Flags - STOP

If you catch yourself:

• Using "should", "probably", "seems to"
• Expressing satisfaction before verification ("Great!", "Perfect!", "Done!")
• About to commit/push/PR without verification
• Trusting subagent success reports
• Relying on partial verification
• Thinking "just this once"
• Tired and wanting work over
• ANY wording implying success without having run verification

ALL of these mean: STOP. Run verification first.

Rationalization Prevention

Key Patterns

Tests:

Copy✅ [Run test command] [See: 34/34 pass] "All tests pass"
❌ "Should pass now" / "Looks correct"

TDD Evidence:

Copy✅ Show failing test output → Show passing test output → "TDD followed"
❌ "I followed TDD" (without evidence)

Spec Compliance:

Copy✅ Show test for Scenario X → Show test for Scenario Y → "Spec compliant"
❌ "All Scenarios have tests" (without showing them)

Build:

Copy✅ [Run build] [See: exit 0] "Build passes"
❌ "Linter passed" (linter doesn't check compilation)

Requirements:

Copy✅ Re-read plan → Create checklist → Verify each → Report gaps or completion
❌ "Tests pass, phase complete"

Subagent delegation:

Copy✅ Subagent reports success → Check actual output → Verify changes → Report actual state
❌ Trust subagent report

SuperSpec Integration

Before Claiming Task Complete

Copy1. Run tests: npm test (or project test command)
2. Show output: "X/Y tests pass"
3. Check Spec coverage: Does test exist for each Scenario?
4. Show evidence: "Test for Scenario X: [test name]"
5. THEN claim: "Task complete"

Before Claiming TDD Followed

Copy1. Show: Failing test output (RED phase)
2. Show: Passing test output (GREEN phase)
3. Verify: Failure came BEFORE implementation
4. THEN claim: "TDD followed"

Before Claiming Phase Complete

Copy1. List all tasks in phase
2. For each task: show completion evidence
3. Run full test suite
4. Show output
5. THEN claim: "Phase complete"

Why This Matters

Trust is earned through evidence:

• Unverified claims break trust
• "I don't believe you" = trust broken
• Undefined functions shipped = would crash
• Missing requirements shipped = incomplete features
• Time wasted on false completion → redirect → rework

Verification is faster than rework:

• 30 seconds to verify
• Hours to debug shipped bugs
• Days to rebuild trust

When To Apply

ALWAYS before:

• ANY variation of success/completion claims
• ANY expression of satisfaction
• ANY positive statement about work state
• Committing, PR creation, task completion
• Moving to next task
• Reporting to user

Rule applies to:

• Exact phrases
• Paraphrases and synonyms
• Implications of success
• ANY communication suggesting completion/correctness

Integration

Called by:

• superspec:execute - Before marking task complete
• superspec:finish-branch - Before presenting options
• superspec:verify - Core verification skill

Pairs with:

• tdd - TDD evidence verification
• All skills that claim completion

The Bottom Line

No shortcuts for verification.

Run the command. Read the output. THEN claim the result.

This is non-negotiable.