Menú
Comprehensive review of Claude Code skills and agents against v0.5 guidelines. Validates metadata, implementation, and architecture. Use when reviewing existing packages or during skill development.
Instalar con Codex o Claude Copia este prompt, pégalo en Codex, Claude u otro asistente, y deja que revise la página de la skill y la instale por ti.
Basado en la clasificación ocupacional SOC
Set up a repo-local just task runner with a root Justfile, optional .just/ helper scripts, and curated help, build, fmt, lint, test, and ci recipes. Use when a repo needs just, a Justfile, .just helpers, or when the user mentions task automation, "just build", "just lint", "just fmt", or dropping in a just system.
Create or harden CLIs intended primarily for AI or system consumption and secondarily for humans. Use when designing or implementing a JSON-first CLI with an MCP-ready contract seam, where every command supports machine output, errors are typed and actionable, mutating commands are auditable via corresponding read commands, and external integrations need simulator-backed testing. Do not use for human-first shell utilities, one-off scripts, or text-only CLIs.
Design stateful simulators for AI-facing CLIs that integrate with devices, networks, services, or databases. Use when a CLI needs realistic simulator-backed testing, configurable fault injection, persistent state, and a swappable adapter boundary so the same business logic runs against live and simulated backends. Do not use for shallow mocks, stateless test doubles, or CLIs with no external integration.
Critically review an existing CLI, MCP wrapper, or CLI plan for AI-first contract quality. Use when evaluating whether a CLI is JSON-first, has an MCP-ready contract seam without JSON reshaping, auditably models state changes, uses stateful simulator-backed testing for external integrations, and returns typed actionable errors. Do not use for general style review or human-first shell UX review.
Run CI quality gates with optional auto-fix and PR creation.
Run Codex tasks via the ai_cli Task Tool runner.
| name | skill-reviewing |
| description | Comprehensive review of Claude Code skills and agents against v0.5 guidelines. Validates metadata, implementation, and architecture. Use when reviewing existing packages or during skill development. |
Provides comprehensive validation of Claude Code skills and agents against:
This skill orchestrates three specialized review agents, each focusing on a distinct concern:
| Agent | Focus | Source Document |
|---|---|---|
skill-metadata-storage-review | Frontmatter & persistence | PLUGIN-STORAGE-CONVENTIONS.md |
skill-implementation-review | Code mechanics & security | agent-tool-use-best-practices.md |
skill-architecture-review | Design patterns & structure | claude-code-skills-agents-guidelines-0.4.md |
All three agents are invoked via Agent Runner for parallel execution:
Validates:
Returns: {success, data: {checks_performed, checks_passed, warnings, errors}, error}
Validates:
Returns: {success, data: {checks_performed, checks_passed, warnings, errors}, error}
Validates:
Returns: {success, data: {checks_performed, checks_passed, warnings, errors}, error}
When user requests a review:
Determine target
Invoke agents in parallel Use Agent Runner to invoke all three reviewers concurrently with parameters:
For skill-metadata-storage-review:
{
"target_type": "<agent|skill|package>",
"target_path": "/absolute/path/to/target",
"package_name": "<package-name-if-applicable>",
"check_registry": true,
"registry_path": ".claude/agents/registry.yaml"
}
For skill-implementation-review:
{
"target_path": "/absolute/path/to/target",
"check_hooks": true,
"check_dependencies": true,
"check_security": true,
"manifest_path": "manifest.yaml"
}
For skill-architecture-review:
{
"target_type": "<agent|skill|package>",
"target_path": "/absolute/path/to/target",
"check_two_tier": true,
"check_contracts": true,
"check_naming": true,
"registry_path": ".claude/agents/registry.yaml"
}
Aggregate results
Format report Present findings to user in this structure:
## Review Summary: [package-name]
✅ Metadata & Storage: X checks passed, Y warnings, Z errors
⚠️ Implementation: X checks passed, Y warnings, Z errors
✅ Architecture: X checks passed, Y warnings, Z errors
### Critical Issues (must fix)
[List all errors with file:line and suggested actions]
### Warnings (should fix)
[List all warnings with suggestions]
### Info (optional improvements)
[List all info-level findings]
Review entire package:
/skill-reviewing sc-managing-worktrees
Review specific agent:
/skill-reviewing .claude/agents/sc-worktree-create.md
Review skill:
/skill-reviewing .claude/skills/managing-worktrees/
Review during development:
Review the skill I just created in .claude/skills/my-new-skill/
.claude/agents/<name>.md.claude/skills/<name>/SKILL.mdpackages/<name>/