CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

claude-plugin-guide

Look up Claude Code plugin authoring contracts in repository references and official documentation; read-only guidance for Claude products.

Quality

63%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.codex/skills/claude-plugin-guide/SKILL.md
SKILL.md
Quality
Evals
Security

Claude plugin authoring reference

Resolve the RU/EN question into the relevant manifest, component, marketplace, hook, or installation topic. Return a concise answer with source paths and supported examples. This is read-only product guidance: do not install, enable, update, launch, or repair plugins, execute Claude commands, or self-update this guide.

Read only references relevant to the question:

  • Native Claude product-authoring rules: distributed Claude skill frontmatter, prompt-first invocation, setup lifecycle, and validation; excludes native Codex skills.
  • Product skill-authoring reference, frontmatter catalog, and prompt contract: canonical tracked skill contracts. Agent-return-setup demonstrates structural frontmatter/PLAN/lifecycle, not default routing: its default descriptions conflict. Validate affected product skill directories with bash brewcode/skills/skills/scripts/validate-skill.sh <affected-skill-directory>.
  • Setup defaults are per skill: compare documented behavior with executable resolution and report disagreements. Teams' parser and docsync use installed status/absent install; semble routing defaults to read-only status, never a bare machine-level install.
  • Product agent-authoring reference: tracked Claude agent contracts, distinct from native .codex/agents/*.toml.
  • Product hook reference: tracked Claude product hook contracts; embedded commands and Claude tools are reference text, never executable Codex instructions.
  • Relevant tracked manifests under each product's .claude-plugin/ and its skills/, agents/, and hooks/: actual implementation. Locate with rg --files; inspect focused ranges.
  • Official plugin reference, marketplace reference, and plugin guide: verify current or uncertain behavior before presenting it as current.

Distinguish tracked implementation, dated reference claims, and verified official behavior. Report source disagreements with evidence. Claude manifests and substitution tokens describe the shipped Claude product; native Codex uses project instructions, skills, agents, and tools with plugins disabled.

Repository

kochetkov-ma/claude-brewcode

Path

.codex/skills/claude-plugin-guide/SKILL.md

Last updated
First committed

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.