skill-test

Validate skill files for structural compliance and behavioral correctness.

Updated Aug 2, 2026
One-click install
npx skills add https://github.com/claretagrapelike32/codex-ai-game-studio --skill skill-test-claretagrapelike32
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-test
Source: https://github.com/claretagrapelike32/codex-ai-game-studio/tree/main/plugins/ai-game-studio/skills/skill-test
Command: npx skills add https://github.com/claretagrapelike32/codex-ai-game-studio --skill skill-test-claretagrapelike32

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Skill authors need a reliable way to verify that their skill packages meet structural requirements and behave correctly when invoked, before shipping them to users. ## Core Features & Use Cases - Static Validation: Lints SKILL.md frontmatter, directory naming, description quality, agents/openai.yaml metadata, and scans for secrets, absolute paths, and platform-specific assumptions. - Behavioral Testing: Runs forward tests with isolated subagents covering explicit invocation, natural-language routing, near-miss rejection, and unsafe request handling. - Use Case: After writing a new skill, run this validator to confirm the frontmatter is compliant, the description triggers correct routing, and safety gates hold under negative prompts, then receive a prioritized improvement proposal. ## Quick Start Use the skill-test skill to validate the skill in my target directory and report a test matrix with any compliance failures.

Frequently Asked Questions about skill-test

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

FAQPage Schema
How do I validate a SKILL.md file for compliance?

Run the static validation mode, which checks that SKILL.md has YAML frontmatter with exactly name and description, that the directory name matches in lowercase hyphen-case, and that relative links resolve. It also runs the official skill validator when available and reports the exit status.

How to test whether a skill activates on the right prompts?

Use forward testing, which spawns isolated subagents with the raw skill instructions. It covers explicit invocation, natural-language routing, near-miss prompts that should route elsewhere, and unsafe requests that must preserve approval gates.

What does the agents/openai.yaml validation check?

It confirms the file has a useful display name, a short description between 25 and 64 characters, and a default prompt that explicitly names the skill. Mutating or destructive workflows must also disable implicit invocation.

Does skill validation modify the tested skill files?

No, testing is read-only by default. The validator reports a test matrix and improvement proposal, and only edits the tested skill after the user separately approves that change.

What security issues does skill static analysis detect?

The scan flags hardcoded secrets, absolute developer paths, path escapes, unsupported model pins, product-specific tool declarations, and platform-only assumptions such as requiring a Unix compatibility layer on Windows.