evaluate

Run tests and review code to decide GO/NO-GO release readiness.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/fractary/claude-plugins --skill evaluate
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: evaluate
Source: https://github.com/fractary/claude-plugins/tree/main/plugins/faber/skills/evaluate
Command: npx skills add https://github.com/fractary/claude-plugins --skill evaluate

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires bash, jq, and includes scripts (resource) components.

What problem does it solve?

Evaluate runs the test suite and reviews the implementation to determine whether to proceed with a release or retry the Build phase.

Core Features & Use Cases

  • Run tests (unit/integration)
  • Review code quality
  • Decide GO/NO-GO with structured results
  • Document failures and inform next steps

Quick Start

Return a GO decision when tests pass, or a NO-GO with detailed failure information for fixes.

Frequently Asked Questions about evaluate

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

FAQPage Schema
How do I run tests and decide if code is ready to release?

Evaluate runs your test suite and reviews code to produce a GO or NO-GO decision. It executes unit, integration, and end-to-end tests, performs linting and security checks, and documents failures so you know exactly what needs fixing before proceeding.

What does a GO/NO-GO decision mean in a release workflow?

A GO decision means tests pass and code meets quality standards—proceed with release. A NO-GO provides detailed failure information and context so you can retry the build phase with specific fixes required.

Can I use Evaluate to check code quality alongside running tests?

Yes. Evaluate combines test execution, code review, linting, and security checks in one step, then returns structured results that capture functional requirements and quality adherence for informed release decisions.

How does Evaluate integrate with build and retry workflows?

Evaluate sits in the Build, Evaluate, and retry loop—it runs after your build completes, assesses readiness, and when NO-GO occurs, the workflow retries the build phase with documented failure details to guide corrections.

What test types does Evaluate support?

Evaluate handles unit, integration, and end-to-end tests alongside spec verification and security checks, giving comprehensive coverage across your test suite to validate release readiness.

Do I need bash and jq to use Evaluate?

Yes. Evaluate depends on bash for script execution and jq for structured result processing, so both must be available in your environment.