test-evidence-review

Evaluate test files and manual evidence for coverage, edge cases, naming, and sign-offs.

Updated Apr 10, 2026
One-click install
npx skills add https://github.com/polsegu/Claude-Code-Game-Dev --skill test-evidence-review-polsegu
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: test-evidence-review
Source: https://github.com/polsegu/Claude-Code-Game-Dev/tree/main/.claude/skills/test-evidence-review
Command: npx skills add https://github.com/polsegu/Claude-Code-Game-Dev --skill test-evidence-review-polsegu

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a structured, evidence-first quality review of automated tests and manual evidence documents to detect gaps that simple existence or smoke checks miss, preventing unverified acceptance criteria from slipping into QA sign-off.

Core Features & Use Cases

  • Holistic test discovery: Locates unit, integration, and UI/visual evidence across tests/, production/ and sprint plans using globbing and grep heuristics.
  • Automated quality heuristics: Counts assertions, checks edge-case references, evaluates test naming, and traces formula references to assess test robustness.
  • Manual evidence auditing: Verifies criterion linkage, sign-off completeness, screenshot/artifact presence, and evidence freshness for Visual/Feel and UI stories.
  • Use Case: Run the skill before QA hand-off to produce per-story ADEQUATE/INCOMPLETE/MISSING verdicts and a conversational summary or optional written report.

Quick Start

Run the test-evidence-review on a story path or sprint to generate a per-story quality verdict and summary report.

Frequently Asked Questions about test-evidence-review

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

FAQPage Schema
How do I evaluate test evidence quality and assertion coverage before QA hand-off?

To evaluate test evidence quality, this skill analyzes test files and manual evidence documents for assertion coverage, edge cases, naming, and sign-offs, producing ADEQUATE, INCOMPLETE, or MISSING verdicts per story before QA hand-off.

What is automated test evidence review and how does it detect coverage gaps?

Automated test evidence review is a process that locates unit, integration, and UI tests using globbing and grep heuristics, then counts assertions and checks edge-case references to detect coverage gaps that simple existence checks miss.

How do I audit manual evidence documents for sprint stories and acceptance criteria?

To audit manual evidence documents, this skill verifies criterion linkage, sign-off completeness, screenshot or artifact presence, and evidence freshness for Visual, Feel, and UI stories within a specified sprint.

Can I run a quality audit across all stories in an epic or just a single story file?

Yes, you can run a quality audit on a single story file, the current sprint, or all stories in a named system or epic, generating per-story assessments and an overall verdict with an optional written report.

What is the best way to trace acceptance criteria to automated tests and prevent unverified sign-offs?

The best way to trace acceptance criteria to automated tests is using heuristic parsing that matches story criteria against test content, preventing unverified acceptance criteria from slipping into QA sign-off.

Does test review work without external dependencies or specific testing frameworks?

Yes, test review operates without external dependencies, using built-in globbing and grep heuristics to locate and parse test content across directories, making it framework-agnostic for assertion coverage analysis.