verification-before-completion
Scannednpx machina-cli add skill faulkdev/github-copilot-superpowers/verification-before-completion --openclawVerification Before Completion
[2026-02-01 CONSOLIDATION NOTE] This is the OBRA version (
obra-verification-before-completion), which was promoted as the primary implementation due to:
- More structured Copilot Completion Checklist (Requirements/Correctness/Quality/Verification)
- Explicit "Why This Matters" context from failure memories
- Agent delegation pattern guidance
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.
Copilot Completion Checklist
✅ Requirements
- Re-state what "done" means (from the user request)
- Confirm each requirement is satisfied (link to files/lines if possible)
✅ Correctness
- Consider edge cases (inputs, nulls, errors)
- Add regression tests when fixing bugs
✅ Quality
- Lint/format follows project rules
- No debug prints/logs left behind unless requested
- Minimal diff; clear naming
✅ Verification commands (choose what fits repo)
- Tests:
… - Lint:
… - Build:
… - Typecheck:
…
❌ No fabricated verification
- If you can't run commands, provide them and state you didn't execute them
The Iron Law
NO 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
BEFORE 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
| Claim | Requires | Not Sufficient |
|---|---|---|
| Tests pass | Test command output: 0 failures | Previous run, "should pass" |
| Linter clean | Linter output: 0 errors | Partial check, extrapolation |
| Build succeeds | Build command: exit 0 | Linter passing, logs look good |
| Bug fixed | Test original symptom: passes | Code changed, assumed fixed |
| Regression test works | Red-green cycle verified | Test passes once |
| Agent completed | VCS diff shows changes | Agent reports "success" |
| Requirements met | Line-by-line checklist | Tests passing |
Red Flags - STOP
- Using "should", "probably", "seems to"
- Expressing satisfaction before verification ("Great!", "Perfect!", "Done!", etc.)
- About to commit/push/PR without verification
- Trusting agent success reports
- Relying on partial verification
- Thinking "just this once"
- Tired and wanting work over
- ANY wording implying success without having run verification
Rationalization Prevention
| Excuse | Reality |
|---|---|
| "Should work now" | RUN the verification |
| "I'm confident" | Confidence ≠ evidence |
| "Just this once" | No exceptions |
| "Linter passed" | Linter ≠ compiler |
| "Agent said success" | Verify independently |
| "I'm tired" | Exhaustion ≠ excuse |
| "Partial check is enough" | Partial proves nothing |
| "Different words so rule doesn't apply" | Spirit over letter |
Key Patterns
Tests:
✅ [Run test command] [See: 34/34 pass] "All tests pass"
❌ "Should pass now" / "Looks correct"
Regression tests (TDD Red-Green):
✅ Write → Run (pass) → Revert fix → Run (MUST FAIL) → Restore → Run (pass)
❌ "I've written a regression test" (without red-green verification)
Build:
✅ [Run build] [See: exit 0] "Build passes"
❌ "Linter passed" (linter doesn't check compilation)
Requirements:
✅ Re-read plan → Create checklist → Verify each → Report gaps or completion
❌ "Tests pass, phase complete"
Agent delegation:
✅ Agent reports success → Check VCS diff → Verify changes → Report actual state
❌ Trust agent report
Why This Matters
From 24 failure memories:
- Your human partner said "I don't believe you" - trust broken
- Undefined functions shipped - would crash
- Missing requirements shipped - incomplete features
- Time wasted on false completion → redirect → rework
- Violates: "Honesty is a core value. If you lie, you'll be replaced."
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
- Delegating to agents
Rule applies to:
- Exact phrases
- Paraphrases and synonyms
- Implications of success
- ANY communication suggesting completion/correctness
The Bottom Line
No shortcuts for verification.
Run the command. Read the output. THEN claim the result.
This is non-negotiable.
Source
git clone https://github.com/faulkdev/github-copilot-superpowers/blob/integrate-obra-superpowers/.github/skills/verification-before-completion/SKILL.mdView on GitHub Overview
Claiming completion without verification is dishonest, not efficient. This skill enforces running verification commands (tests, lint, build, typecheck) and citing verifiable outputs before asserting success.
How This Skill Works
Identify the command that proves the claim (tests, lint, build, typecheck). Run the FULL command in a fresh context. Read the full output and exit code, then verify whether the results substantiate the claim; only then state status with evidence.
When to Use It
- Before committing or pushing changes
- Before opening a pull request or status update
- After implementing a feature or bug fix and running verification commands
- When reporting success to teammates or documenting progress
- Before marking a milestone as done in release notes or project docs
Quick Start
- Step 1: Identify the verification command that proves the claim (tests, lint, build, typecheck)
- Step 2: Run the FULL command in a fresh environment and collect outputs
- Step 3: Read the outputs, confirm exit codes and failures, then claim success only with evidence
Best Practices
- Restate what 'done' means from the user request and verify each requirement (link to files/lines if possible)
- Consider edge cases and add regression tests for fixes
- Enforce quality: lint/formatting rules, remove debug prints/logs, keep diffs minimal and naming clear
- Use explicit verification commands (tests, lint, build, typecheck) and rely on fresh outputs
- If you cannot run commands, provide the commands and clearly state you didn’t execute them
Example Use Cases
- Run the test suite and confirm 0 failures, then claim tests pass with evidence from the test runner
- Execute lint with 0 errors and include the lint output as verification
- Perform a full build that exits cleanly (exit code 0) and cite the build log
- Validate a bug fix with a regression test that reproduces the original symptom and passes after the fix
- Document completion only after obtaining fresh verification outputs; avoid claiming success without them