Skip to main content
affpapa.org
RU EN
skill Claude Code CX Codex Cursor GitHub Copilot

verify · yeachan-heo — Claude Code skill: what it does, how to install

Turns "it should work" into evidence you can read: which check ran, what passed, and what stayed unverified.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓ 
npx skills add Yeachan-Heo/oh-my-claudecode --skill verify -a claude-code
Open on GitHub

What verify does

Reach for this skill when someone asks whether a change really works and the answer so far is a shrug. It turns that into evidence: which behaviour had to be proven, which command proved it, and what came back.

Verification runs in a set order. Existing tests come first, then typecheck and build, then the narrowest direct command that hits the behaviour, and manual or interactive checks last. When coverage is missing, the skill hunts for the smallest direct check instead of inventing a test suite on the spot.

The file is a single SKILL.md holding the workflow and the rules. Saying a change is done without evidence breaks the first rule, and a failing check has to appear in the report as it happened. If there is no realistic way to verify something, the skill says so outright rather than bluffing. The report covers four things: what was verified, which commands and tests ran, what passed, what failed or stayed unverified.

What it helps with

  • prove a fix closes the bug instead of merely compiling
  • pick a check to run when the new code has no tests
  • report on a refactor without a vague "works for me"
  • summarise a verification run without dumping raw logs
  • name the parts that stayed unverified

Example prompts for the agent

  • Verify that the status change actually works: run the existing tests and show me the output.
  • I finished the refactor of `src/auth/session.ts`. Prove that sign-in and sign-out still work.
  • If there's no way to check this, say plainly what stays unverified.

How to install verify

1Skills CLI (Vercel)

npx skills add Yeachan-Heo/oh-my-claudecode --skill verify -a claude-code

2Manually: put the files in place

~/.claude/skills/verify/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/verify/

From the same set26

The whole set →
ai-slop-cleaner Yeachan-Heo Cleans bloated AI-generated code by deleting dead logic and duplicate abstractions. It keeps current behavior locked under regression tests. Claude Code CX Codex Cursor +1 1,500installs ralplan Yeachan-Heo A skill that runs iterative planning with three agents (Planner, Architect, Critic) until consensus. Blocks vague requests before execution, never runs changes without your approval. Claude Code CX Codex Cursor +1 1,200installs hud Yeachan-Heo The hud skill configures a HUD display in Claude Code's statusline, showing git branch, context usage, running subagents, and other metrics. Track session state and keep tabs on complex work without losing context. Claude Code CX Codex Cursor +1 1,100installs deepinit Yeachan-Heo Walks a repository and writes an AGENTS.md per directory: what sits there, how the parts connect, what an agent should know before editing. Claude Code CX Codex Cursor +1 1,000installs autopilot Yeachan-Heo A skill that takes a task description and runs the full development cycle: requirements analysis, design, implementation, testing, and validation—all hands-off. Use it when you want to describe what you need and get working code back. Claude Code CX Codex Cursor +1 999installs deep-interview Yeachan-Heo A skill that runs a Socratic interview on your idea, scores ambiguity across weighted dimensions, and blocks coding until the score falls under the threshold. Claude Code CX Codex Cursor +1 998installs omc-setup Yeachan-Heo Canonical setup and repair wizard for oh-my-claudecode. It configures CLAUDE.md files, HUD statusline, and agent team defaults. Claude Code CX Codex Cursor +1 987installs ralph Yeachan-Heo A persistence loop that works through tasks in prd.json until all stories pass and get reviewer verification. Useful for multi-step jobs that need guaranteed completion. Claude Code CX Codex Cursor +1 971installs

Other versions of verify

verify react Validates code changes against React repository requirements before committing. Runs Prettier, linter, Flow type checks, and tests. Claude Code CX Codex Cursor +1 1,200installs

Similar

observe-metrics ruvnet Aggregates agent swarm counters, gauges, and histograms over a time window to detect metric anomalies against baselines. Claude Code CX Codex Cursor +1 733installs sparc-spec ruvnet Runs the first SPARC phase: turns a feature description into requirements, acceptance criteria and constraints, then keeps the spec in ruflo memory. Claude Code CX Codex Cursor +1 732installs observe-trace ruvnet Collects distributed trace spans for a task and builds an execution tree. It highlights agent timing, bottlenecks, and the critical path across workflows. Claude Code CX Codex Cursor +1 732installs vector-embed ruvnet Turns a string or a file into a 384-dimension vector with ruvector and keeps it in an HNSW index for semantic search. Claude Code CX Codex Cursor +1 731installs trader-risk ruvnet Evaluates portfolio VaR, CVaR, drawdown limits, and circuit breaker status using the neural-trader risk engine. Claude Code CX Codex Cursor +1 731installs trader-regime ruvnet The skill asks neural-trader which market regime is running and suggests a strategy type to match. Handy before a backtest or a signal. Claude Code CX Codex Cursor +1 731installs setup-matt-pocock-skills Matt Pocock This skill configures a repository for Matt Pocock's engineering skills by setting its issue tracker, triage labels, and domain docs layout. Run it before using the other skills. Claude Code 1,052,851installs to-tickets Matt Pocock Turns a plan, spec, or conversation into small end-to-end tickets with explicit blockers. Use it to agree on the work breakdown before tickets go to a tracker. Claude Code 711,230installs

Questions about verify

How do I install verify in Claude Code?

Run npx skills add Yeachan-Heo/oh-my-claudecode --skill verify -a claude-code. The skill lands in ~/.claude/skills/verify/SKILL.md.

What does the verify skill do?

It makes the agent prove a change works before calling the work complete, and report only what it actually checked.

What order does it verify in?

Existing tests first, then typecheck and build, then narrow direct command checks, and manual or interactive validation last.

Source text

SKILL.md · MIT 
---
name: verify
description: Verify that a change really works before you claim completion
---

# Verify

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

## Verification order
1. Existing tests
2. Typecheck / build
3. Narrow direct command checks
4. 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

© Yeachan-Heo · MIT · Yeachan-Heo/oh-my-claudecode

See also

Project sponsors

Companies that keep this analytics open
Gold sponsor
Nutra.Club

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor