It makes an agent test that a change actually works before declaring a task done, so completion is based on verification rather than a claim.

When should it be used?

Whenever an agent makes a change and you want confirmation that the change really works before the task is treated as finished.

Verify

SkillAI & models

Verify is a skill that makes an AI agent test that a change actually works before declaring a task done. Instead of accepting an agent's claim that something is finished, it verifies the change really works first. This reduces the chance of a task being marked complete when it is not.

Use Verify in Claude, ChatGPT or Ahel Desktop

Free. Sign in, add Verify and connect your AI. About a minute.

Claude ChatGPT Ahel Desktop

Also: Claude Code · Cursor · Codex

Then ask your AI: use the Verify skill

Details

Instructions available. Your AI can read the instructions. Execution depends on the setup they require.

Account requirements not reviewed. Check the skill instructions before use; ahel provides instructions and does not run this skill.

No other account needed.

Have an AI agent setup that supports skills.

VerifyStart free

What your AI can do with it

Test that a change actually works before declaring a task done
Verify a change really works before completion is claimed
Stop an agent from marking a task complete without testing

Getting started

Have an AI agent setup that supports skills.
Add the verify skill to that setup.
Ask the agent to use verify when working on changes so it tests them before claiming completion.

What this skill tells your AI

The instructions your AI receives, as published by yeachan-heo/oh-my-claudecode in skills/verify/SKILL.md and read by ahel’s review.

Use this skill when the user wants confidence that a feature, fix, or refactor actually works.

Goal

Turn vague “it should work” claims into concrete evidence.

Workflow

Identify the exact behavior that must be proven.
Prefer existing tests first.
If coverage is missing, run the narrowest direct verification commands available.
If direct automation is not enough, describe the manual validation steps and gather concrete observable evidence.
Report only what was actually verified.

Verification order

Existing tests
Typecheck / build
Narrow direct command checks
Manual or interactive validation

Rules

Do not say a change is complete without evidence.
If a check fails, include the failure clearly.
If no realistic verification path exists, say that explicitly instead of bluffing.
Prefer concise evidence summaries over noisy logs.

Output

What was verified
Which commands/tests were run
What passed
What failed or remains unverified

Signals

GitHub stars: 39k
Forks: 4k
Last commit: Oct 2026

Questions

What does verify do?: It makes an agent test that a change actually works before declaring a task done, so completion is based on verification rather than a claim.
When should it be used?: Whenever an agent makes a change and you want confirmation that the change really works before the task is treated as finished.

Advanced

Item type: skill
Key: verify-yeachan-heo
Source: github.com/yeachan-heo/oh-my-claudecode

github.com/yeachan-heo/oh-my-claudecode

More in AI & models

skill-creator
Skill · anthropics
More in AI & models wayfinder
Skill · mattpocock
More in AI & models wizard
Skill · mattpocock
More in AI & models algorithmic-art
Skill · anthropics
More in AI & models code-review-and-quality
Skill · addyosmani
More in AI & models ai-first-engineering
Skill · affaan-m
More in AI & models