check-work

Verify completed changes with fresh evidence before claiming completion.

3|1|Updated May 10, 2026
One-click install
npx skills add https://github.com/nuttaruj/rolepod --skill check-work
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: check-work
Source: https://github.com/nuttaruj/rolepod/tree/main/core/skills/check-work
Command: npx skills add https://github.com/nuttaruj/rolepod --skill check-work

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill prevents premature completion claims by forcing fresh, concrete verification after code, UI, config, docs, or other project changes. It helps teams replace guesswork and stale assumptions with reproducible evidence, while making any verification gaps and risks explicit.

Core Features & Use Cases

  • Evidence-first verification: Runs the right proof for the change type, such as tests, builds, typechecks, curl checks, browser observation, or screenshots.
  • False-green prevention: Enforces strong assertion discipline, red-green-revert validation for bug fixes, and spec-to-evidence traceability so passing checks actually prove the intended behavior.
  • Cross-phase handoff support: Validates subagent or plugin output, aggregates child evidence manifests, and routes failures back to debugging or implementation when proof does not hold.
  • Use cases: Verify a bug fix with regression evidence, confirm a UI change through browser observation, validate API behavior with command output, or document limitations when infrastructure prevents a full check.

Quick Start

After making a change, use check-work to run fresh verification commands, capture the exact proof lines, and produce an evidence block before saying the work is done.

Frequently Asked Questions about check-work

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I prevent false-green test results when validating a bug fix?

False-green prevention enforces strong assertion discipline and spec-to-evidence traceability. It requires red-green-revert validation for bug fixes to ensure passing regression checks actually prove the intended behavior.

How does subagent verification work for cross-phase handoff support?

Cross-phase handoff support validates subagent output and aggregates child evidence manifests. It routes failures back to debugging or implementation when verification proof does not hold.

How do I validate API behavior with command output after schema changes?

You can validate API behavior by running exact verification commands and capturing command output as proof. The process requires concrete evidence like curl checks to confirm API updates function as intended before completion.

What is spec-to-evidence traceability and how does it map acceptance criteria?

Spec-to-evidence traceability maps each acceptance criterion to specific verification evidence. This ensures every required behavior has corresponding proof, making verification gaps explicit before completion claims are made.