$ loading_
Verify results before claiming work is complete, fixed, or passing.
This skill appears to be a prompt-only workflow guideline with no required secrets, no declared remote endpoints, and no described file or code execution capability. Based on the available materials and its open-source source, overall risk is low, with the main caveat being limited adoption and unclear maintenance.
The materials explicitly state that no keys or environment variables are required; the README also does not ask for login, token injection, or credential passing, so no clear credential leakage or abuse surface is evident.
Both the system checks and the materials indicate no remote endpoints; this is a prompt-style ruleset and does not describe sending user data to external services or performing network validation.
Although the README discusses 'running verification commands,' the audited artifact itself is marked prompt-only and does not provide an executable, install script, or local process-launching mechanism; the material describes behavioral rules rather than code execution capability.
It does not declare any need to read or write local files, repository contents, system resources, or external data sources; from the materials, it only provides a pre-completion verification guideline and does not involve additional data access permissions.
Positive factors include being open-source on GitHub and auditable source code; however, the license is unspecified, community adoption is 0 stars, and maintenance status is unknown, so supply-chain trust is moderate and warrants manual review of repository contents and commit history.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "verification-before-completion" skill from askskill: 1. Download https://raw.githubusercontent.com/obra/superpowers/main/skills/verification-before-completion/SKILL.md 2. Save it as ~/.claude/skills/verification-before-completion/SKILL.md 3. Reload skills and tell me it's ready
I just fixed an error in the login API. Do not claim it is fixed yet. First list the verification commands to run, explain what output to check at each step, and only judge whether it is truly fixed after I share the results.
It provides verification steps, pass criteria, and required evidence before deciding whether the fix can be claimed successful.
Before creating a PR, help me identify which build, test, and lint commands should be run for this project. Only conclude that it is ready for a PR if those command outputs meet expectations.
It outputs a pre-PR verification checklist and stresses that conclusions must be based on command results, not assumptions.
I think the data cleaning script is working now. First design a verification plan including run commands, key logs, sample results, and anomaly checks; do not say the task is complete before seeing that evidence.
It gives data-pipeline verification steps and acceptance criteria, ensuring evidence is reviewed before confirming completion.
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.
NO COMPLETION CLAIMS WITHOUT FRESH VERIFICATION EVIDENCE
If you haven't run the verification command in this message, you cannot claim it passes.
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
| 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 |
| 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 |
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
From 24 failure memories:
ALWAYS before:
Rule applies to:
No shortcuts for verification.
Run the command. Read the output. THEN claim the result.
This is non-negotiable.
Execute implementation plans by splitting and advancing independent tasks in-session.
Systematically investigate bugs, test failures, and unexpected behavior before fixing.
Helps decide merge, PR, or cleanup steps after branch work is complete.
Turn requirements into a clear step-by-step execution plan before implementation.
Set conversation rules to discover and invoke skills before replying.
Clarify intent, requirements, and solution direction before any creative implementation work.
Run verification checks before completion to ensure results are truly correct.
Validate code changes before commit and check React contribution requirements.
Verify AI agent completion claims by checking output file existence and freshness.
Provides systematic verification for Claude Code sessions to improve correctness and quality.
Run a pre-release verification loop for Quarkus builds, tests, scans, and reviews.
Verify claims against evidence as supported, refuted, or not enough evidence.