cc-plugins

Investigate a scope for problems and opportunities. Use when investigating a codebase, reviewing a feature, auditing configuration, assessing for problems, or exploring and reporting findings.

Manage project standards compliance — audit existing projects or scaffold new ones. Triggers: "audit this project", "project audit", "check compliance", "new project", "scaffold project", "project setup".

Use when iterating through findings, working through review items, processing improvements one by one, or implementing changes from an assessment.

MANDATORY load before writing or editing any eval scenario (scenario.yaml), check (checks.yaml), or evals.yaml — wherever the eval suite lives. Evaluate Claude Code configurations (skills, CLAUDE.md, hooks, MCP servers, settings, sub-agents) using the scuttlerun/pincenez/craboodle eval pipeline. TRIGGER when: authoring or modifying scenario.yaml or checks.yaml files; designing checks; interpreting eval results; benchmarking configurations; testing CLAUDE.md effectiveness; verifying hooks work; regression testing configs; building or extending an eval suite. DO NOT skip because 'I'll consult check-design.md directly' or 'one quick check edit' — load first, the schema and anti-pattern rules are cheap to surface.

MANDATORY load before any Read/Edit/Write to plugin content — plugin.json, marketplace.json, files under commands/, agents/, skills/, hooks/, or .mcp.json — regardless of where the plugin lives (personal ~/.claude/plugins/, marketplace-installed, or project-local). Creates, audits, ships, and maintains Claude Code plugins; manages marketplace entries; validates publish readiness. TRIGGER when: editing/creating/reviewing plugin.json or marketplace.json; creating a new plugin; auditing or improving a plugin; regenerating a marketplace; preparing a plugin to publish; bulk-auditing every installed plugin; bootstrapping evals for a plugin. DO NOT skip because "it's only a manifest tweak" — load first, lightweight checks are cheap.

MANDATORY load before any Read/Edit/Write to skill content — SKILL.md, files under scripts/, workflows/, references/, evals/, assets/, or provenance.yml — regardless of where the skill lives (personal ~/.claude/skills/, plugin-installed under ~/.claude/plugins/, or project-local .claude/skills/). Creates, audits, improves, and updates Claude Code skills; integrates content from external sources. TRIGGER when: editing/creating/reviewing any file inside a skill directory; creating a new skill; auditing or improving a skill; bootstrapping evals for a skill; integrating or folding in an external source; updating skills from upstream. DO NOT skip because "it's only a script change" or "only a workflow tweak" — load first, lightweight checks are cheap.