Verification Loop Runner skill for Claude and Codex

Verification Loop Runner as an AI skill. It asks for The task, Repository checks available, Test scope per change and 3 more, then does the rest. Free to…

Install it once and Claude or Codex runs it for you. When your request matches, your AI loads the skill, asks for anything it is missing, and works through the template itself. Prefer to fill it in by hand? See the full prompt template.

Download the skill (.zip) · View SKILL.md

When it is used

Use when a multi-step code task needs each change followed by its tests and checks, with evidence recorded before the next change starts. Do not use for writing the tests themselves or planning the architecture of a feature. Related topics: coding, productivity.

What it will ask you for

• The task (longer free text)
• Repository checks available (longer free text)
• Test scope per change (choose one: Only affected tests, Affected plus integration, Full suite)
• Where to record evidence (choose one: Reply in chat, A progress file, Commit messages)
• Stop on first failure (choose one: Yes, No)
• Definition of done (longer free text)

How it works

1. Split the task into the smallest changes that can each be checked on their own.
2. Before each edit, state the check that will prove the change, such as a failing test or an observable behaviour.
3. Make the change, run the chosen scope of checks, and record the exact command and result.
4. Fix any failure before starting the next change, so work never stacks on a red result.
5. Run the full set of checks once at the end and report done or not done against the definition of done.

What you get

A table with one row per change showing the check run, result, and evidence, followed by a plain done or not done statement and any check that could not run.

What it watches out for

• Running only the new test and missing a regression elsewhere that the broader scope would catch.
• Recording that checks passed without the command output, which leaves nothing to verify later.
• Batching several edits before running anything, so a failure cannot be traced to one change.

How to install

Claude.ai and the Claude desktop app

1. Download the skill .zip.
2. Open Settings, then Capabilities, and find Skills.
3. Choose Upload skill and select the .zip.
4. Ask Claude for help with the task. It loads the skill when your request matches.

Claude Code

1. Download the skill .zip.
2. Unzip it into ~/.claude/skills/ (or .claude/skills/ inside a project).
3. Start a new session. Claude Code loads the skill when your request matches.

Codex

1. Download the skill .zip.
2. Unzip it into ~/.agents/skills/ (or .agents/skills/ inside a project).
3. Codex picks it up on its own. Type $ and the skill name to use it, or just ask for the task.

Cursor

1. Download the Cursor rule (.mdc).
2. Save it in .cursor/rules/ inside your project.
3. Cursor's agent applies the rule when your request matches its description.

AGENTS.md (Codex, Cursor, and other coding agents)

1. Copy the AGENTS.md section.
2. Paste it into AGENTS.md at the root of your repository, creating the file if needed.
3. Coding agents that read AGENTS.md follow it when your request fits.