shipyard-verification

Run verification commands and inspect exit statuses before completion claims.

66|3|Updated Jan 31, 2026
One-click install
npx skills add https://github.com/lgbarn/shipyard --skill shipyard-verification
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: shipyard-verification
Source: https://github.com/lgbarn/shipyard/tree/main/skills/shipyard-verification
Command: npx skills add https://github.com/lgbarn/shipyard --skill shipyard-verification

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill ensures completion claims are always backed by fresh verification results, preventing premature or dishonest status updates.

Core Features & Use Cases

  • Fresh verification before completion: run the full verification command and validate results prior to making any success claims.
  • Evidence-first reporting: produce verifiable outputs (logs, exit codes) to justify status decisions.
  • Gatekeeping for commits/PRs: block shipping actions until verification passes.

Quick Start

Use the verification workflow to test the current work state and report findings before asserting success.

Frequently Asked Questions about shipyard-verification

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

FAQPage Schema
How do I require fresh verification evidence before a pull request or commit?

To require fresh verification evidence before a commit or pull request, you apply a quality gate that runs the full verification command and inspects the output and exit status to confirm the true work state.

What is evidence-first reporting for shipping decisions?

Evidence-first reporting is the process of producing verifiable outputs, such as logs and exit codes, to justify status decisions and prevent premature or dishonest completion claims before shipping.

How do I block shipping actions until verification passes?

You can block shipping actions by applying a gatekeeping workflow that requires running the full verification command and validating its results before any success claims or commits are allowed.

Why do I need to run the full verification command before claiming completion?

Running the full verification command is necessary to determine the true state of the work by inspecting actual output and exit status, ensuring completion claims are always backed by fresh verification results.

Can I use this verification workflow to prevent premature status updates?

Yes, you can use this verification workflow to test the current work state and report findings before asserting success, effectively preventing premature or dishonest status updates.

Does this quality gate work without any external dependencies?

Yes, this quality gate operates without external dependencies, requiring only the execution of the full verification command and inspection of its output and exit status to validate the work state.