verification-before-completion

Force verification before claiming success or completion. Prevents false "it works" claims. Triggers when about to say "done", "complete", "works", "fixed", or "the implementation is ready". Requires actually running builds/tests and showing output before claiming success.

25 stars

Best use case

verification-before-completion is best used when you need a repeatable AI agent workflow instead of a one-off prompt.

Force verification before claiming success or completion. Prevents false "it works" claims. Triggers when about to say "done", "complete", "works", "fixed", or "the implementation is ready". Requires actually running builds/tests and showing output before claiming success.

Teams using verification-before-completion should expect a more consistent output, faster repeated execution, less prompt rewriting.

When to use this skill

  • You want a reusable workflow that can be run more than once with consistent structure.

When not to use this skill

  • You only need a quick one-off answer and do not need a reusable workflow.
  • You cannot install or maintain the underlying files, dependencies, or repository context.

Installation

Claude Code / Cursor / Codex

$curl -o ~/.claude/skills/verification-before-completion/SKILL.md --create-dirs "https://raw.githubusercontent.com/ComeOnOliver/skillshub/main/skills/aiskillstore/marketplace/ancplua/verification-before-completion/SKILL.md"

Manual Installation

  1. Download SKILL.md from GitHub
  2. Place it in .claude/skills/verification-before-completion/SKILL.md inside your project
  3. Restart your AI agent — it will auto-discover the skill

How verification-before-completion Compares

Feature / Agentverification-before-completionStandard Approach
Platform SupportNot specifiedLimited / Varies
Context Awareness High Baseline
Installation ComplexityUnknownN/A

Frequently Asked Questions

What does this skill do?

Force verification before claiming success or completion. Prevents false "it works" claims. Triggers when about to say "done", "complete", "works", "fixed", or "the implementation is ready". Requires actually running builds/tests and showing output before claiming success.

Where can I find the source code?

You can find the source code on GitHub using the link provided at the top of the page.

SKILL.md Source

# Verification Before Completion

Force verification before claiming success or completion. Prevents false "it works" claims.

## Purpose

Claude often claims things "work" or are "complete" without actually verifying. This skill ensures
actual verification happens before any success claim.

## Triggers

Activate this skill when you're about to say ANY of:

- "Done"
- "Complete"
- "Finished"
- "Works"
- "Fixed"
- "The implementation is ready"
- "This should work"
- "I've implemented..."

## NEVER Claim Success Without

### For Code Changes

- [ ] `dotnet build` passes (no errors)
- [ ] `dotnet test` passes (or explicit reason why skipped)
- [ ] No new warnings introduced
- [ ] Actually ran the code and showed output

### For Factual Claims

- [ ] Verified against assertions.yaml
- [ ] WebSearch if claim involves dates/versions/status
- [ ] Source cited

### For "It Works" Claims

- [ ] Actually executed the code
- [ ] Showed the output to user
- [ ] Tested at least one edge case

### For "Bug Fixed" Claims

- [ ] Reproduced the original bug
- [ ] Applied the fix
- [ ] Verified bug no longer occurs
- [ ] Ran regression tests

## Red Flag Thoughts (REJECT THESE)

| If you're about to say... | Stop and... |
|---------------------------|-------------|
| "This should work" | Actually test it |
| "The fix is complete" | Run the build |
| "I've implemented the feature" | Show test output |
| "The bug is fixed" | Reproduce - fix - verify |
| ".NET 10 preview supports..." | WebSearch to verify version |
| "This is the correct approach" | Cite source or run tests |

## Output Format

Before ANY completion claim, show:

```markdown
## Verification Checklist

### Build Status

- [ ] `dotnet build` - [PASS/FAIL]
- [ ] `dotnet test` - [PASS/FAIL/SKIPPED: reason]

### Verification Steps Performed

1. [What you actually did]
2. [What output you saw]
3. [What edge case you tested]

### Confidence

[high/medium/low] - [why]

### Ready to Claim Complete

[YES/NO] - [if NO, what's missing]
```

## Anti-Pattern Examples

**WRONG:**

```text
"I've fixed the null reference bug. The implementation is complete."
[No build run, no test run, no verification shown]
```

**CORRECT:**

```text
"Let me verify the fix...

$ dotnet build
Build succeeded.

$ dotnet test --filter "NullReferenceTests"
Passed! 3/3 tests passing.

Verification complete. The null reference bug is fixed."
```

**WRONG:**

```text
"This React component should render correctly now."
```

**CORRECT:**

```text
"Let me verify...

$ npm run dev
[Shows browser screenshot or console output]

Component renders correctly. Tested:
- Initial load
- Empty state
- Error state

Verification complete."
```

## Integration

This skill is the FINAL checkpoint before claiming completion:

```text
[all other skills] - analysis complete
     |
[implementation] - code written
     |
[verification-before-completion] - THIS SKILL
     |
     +-- Run build
     +-- Run tests
     +-- Show output
     +-- THEN claim success
     |
[Present to user] - "Ready to proceed?"
```

## The Golden Rule

> **If you didn't run it, you don't know if it works.**
> **If you didn't verify it, don't claim it's complete.**

Related Skills

cursor-tab-completion

25
from ComeOnOliver/skillshub

Master Cursor Tab autocomplete, ghost text, and AI code suggestions. Triggers on "cursor completion", "cursor tab", "cursor suggestions", "cursor autocomplete", "cursor ghost text", "cursor copilot".

viral-verification

25
from ComeOnOliver/skillshub

验证文章是否符合爆款推文要素,并提供修改建议和评分。

verification-hygiene

25
from ComeOnOliver/skillshub

External evidence discipline and search execution routing. Bridges structure_judgment and judgment_hygiene to govern how the model searches, what it retrieves, when to stop, and how to format evidence before internal reasoning. Prevents treating SEO-driven internet as infallible.

when-verifying-quality-use-verification-quality

25
from ComeOnOliver/skillshub

Comprehensive quality verification and validation through static analysis, dynamic testing, integration validation, and certification gates

verification-quality-assurance

25
from ComeOnOliver/skillshub

Comprehensive truth scoring, code quality verification, and automatic rollback system with 0.95 accuracy threshold for ensuring high-quality agent outputs and codebase reliability.

verification-protocol

25
from ComeOnOliver/skillshub

Independent verification of task completion - eliminates self-attestation

evidence-verification

25
from ComeOnOliver/skillshub

This skill teaches agents how to collect and verify evidence before marking tasks complete. Inspired by production-grade development practices, it ensures all claims are backed by executable proof:...

completion-integrity

25
from ComeOnOliver/skillshub

Prevents shortcuts and cheating when completing tasks. Blocks commits with warning suppressions, commented tests, or deleted assertions.

verification-loop

25
from ComeOnOliver/skillshub

A comprehensive verification system for Claude Code sessions.

springboot-verification

25
from ComeOnOliver/skillshub

Verification loop for Spring Boot projects: build, static analysis, tests with coverage, security scans, and diff review before release or PR.

Product Lens — Think Before You Build

25
from ComeOnOliver/skillshub

## When to Use

laravel-verification

25
from ComeOnOliver/skillshub

Verification loop for Laravel projects: env checks, linting, static analysis, tests with coverage, security scans, and deployment readiness.