Explain the gbrain skillpack contract and required artifacts for validation.
npx skills add https://github.com/weiping/gbrain-cn --skill reference-pack
Please help me install this Agent Skill. Skill: reference-pack Source: https://github.com/weiping/gbrain-cn/tree/main/examples/skillpack-reference/skills/reference-pack Command: npx skills add https://github.com/weiping/gbrain-cn --skill reference-pack
Users need a clear, canonical reference for how third-party gbrain skillpacks are structured and what artifacts must exist for the skillpack contract to be valid and publishable.
Ask your agent to explain the skillpack contract, and it will direct you to the required artifacts inside the pack.
High-intent search queries and answers about installing and using this skill.
The 10/10 skillpack contract specifies the four required artifacts every third-party pack must ship: skillpack.json metadata, SKILL.md frontmatter and body, bootstrap runbooks, and CHANGELOG version entries.
You validate skillpack artifacts by applying the doctor rubric, which scores required metadata, routing files, and CHANGELOG version entries to ensure your pack passes publish-time validation checks.
Skill routing relies on SKILL.md frontmatter and routing-eval intent coverage to map user requests to the correct skill triggers, ensuring agents direct queries to the appropriate skillpack behaviors.
Required files include skillpack.json metadata, skills/<name>/SKILL.md with YAML frontmatter and body, runbooks/bootstrap.md steps, and CHANGELOG version entries to satisfy the skillpack contract.
The doctor rubric fails validation when skillpacks are missing required metadata, YAML frontmatter, bootstrap runbook steps, or CHANGELOG version entries needed to satisfy the skillpack contract.
Yes, YAML frontmatter in skills/<name>/SKILL.md is required for skillpack discovery and routing, as it provides the metadata agents use to map user requests to the correct skill triggers.
inbrainfunDescribe the four core artifacts of third-party inbrain skillpacks.
brianruggieriRun blind extraction evaluations against fixtures and grade results with a rubric.
matteocervelliGenerate CHANGELOG entries from commits using conventional commits and semantic versioning.
openwopValidate OpenWOP protocol changes against non-functional requirements.
