Skip to content

testing.md

Latest commit

 

History

History
29 lines (18 loc) · 1.69 KB

testing.md

File metadata and controls

29 lines (18 loc) · 1.69 KB

Testing and Validation

This is a pure-content repository — JSON manifests + markdown skills. There is no build step, package manager, or automated test suite. Testing means installing the marketplace locally and validating structure and skill quality.

Local Testing

Install the marketplace from your working copy (inside a Claude Code session):

/plugin marketplace add /path/to/ai-plugins
/plugin install openehr-specs@openehr

Then exercise the skills against a checkout of a specifications-XX repo: confirm each skill triggers on its documented phrases (see the description frontmatter) and does not trigger on its anti-trigger cases.

After editing skill content, reinstall (or restart the session) to pick up changes.

For Cursor, add your working copy as a local plugin marketplace and reload the plugin after changes — see install.md.

Validation

  • Manifest/skill validation — python3 scripts/validate.py (also run by CI on every PR): checks Claude and Cursor JSON manifests, name/version sync, declared component paths, marketplace owner.name, and SKILL.md frontmatter.
  • Structural validation — after creating or modifying plugin components, run the plugin-dev:plugin-validator agent: checks plugin.json, marketplace entry, directory layout, and frontmatter.
  • Skill quality review — run the plugin-dev:skill-reviewer agent: checks description triggering quality, progressive disclosure, and content structure.
  • Token cost — claude plugin details openehr-specs shows the component inventory and projected token cost; keep skill metadata lean.

Releasing

See versioning.md for the semver policy and release steps.