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.
Best use case
verification-before-completion is best used when you need a repeatable AI agent workflow instead of a one-off prompt. It is especially useful for teams working in multi. 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.
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.
Users should expect a more consistent workflow output, faster repeated execution, and less time spent rewriting prompts from scratch.
Practical example
Example input
Use the "verification-before-completion" skill to help with this workflow task. Context: 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.
Example output
A structured workflow result with clearer steps, more consistent formatting, and an output that is easier to reuse in the next run.
When to use this skill
- Use this skill when you want a reusable workflow rather than writing the same prompt again and again.
When not to use this skill
- Do not use this when you only need a one-off answer and do not need a reusable workflow.
- Do not use it if you cannot install or maintain the related files, repository context, or supporting tools.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/verification-before-completion/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How verification-before-completion Compares
| Feature / Agent | verification-before-completion | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/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
viral-verification
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
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
Comprehensive quality verification and validation through static analysis, dynamic testing, integration validation, and certification gates
verification-quality-assurance
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
Independent verification of task completion - eliminates self-attestation
evidence-verification
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
Prevents shortcuts and cheating when completing tasks. Blocks commits with warning suppressions, commented tests, or deleted assertions.
azure-quotas
Check/manage Azure quotas and usage across providers. For deployment planning, capacity validation, region selection. WHEN: "check quotas", "service limits", "current usage", "request quota increase", "quota exceeded", "validate capacity", "regional availability", "provisioning limits", "vCPU limit", "how many vCPUs available in my subscription".
raindrop-io
Manage Raindrop.io bookmarks with AI assistance. Save and organize bookmarks, search your collection, manage reading lists, and organize research materials. Use when working with bookmarks, web research, reading lists, or when user mentions Raindrop.io.
zlibrary-to-notebooklm
自动从 Z-Library 下载书籍并上传到 Google NotebookLM。支持 PDF/EPUB 格式,自动转换,一键创建知识库。
discover-skills
当你发现当前可用的技能都不够合适(或用户明确要求你寻找技能)时使用。本技能会基于任务目标和约束,给出一份精简的候选技能清单,帮助你选出最适配当前任务的技能。
web-performance-seo
Fix PageSpeed Insights/Lighthouse accessibility "!" errors caused by contrast audit failures (CSS filters, OKLCH/OKLAB, low opacity, gradient text, image backgrounds). Use for accessibility-driven SEO/performance debugging and remediation.