Menú
Guide for creating valid Claude Code skills. Use when user asks how to create a skill, make a skill, or needs help with skill structure.
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
Use the codex MCP server to review the current git diff. Focus on correctness, regressions, security issues, tests, and maintainability. Return only actionable findings with file/line references and severity. Do not modify files.
Use when the user asks to fetch / list / read / show unresolved CodeRabbit review threads on a GitHub PR. Pulls thread URL, file:line, author, and full comment body for every thread CodeRabbit opened that is not yet resolved, plus CodeRabbit's review-summary nitpicks.
Use when the user asks to process / triage / address / handle PR reviews from CodeRabbit, SonarQube, Codex, and CI all together. First waits (polling every 1 min, max 30 min) until CodeRabbit, Sonar, and every CI check on the PR are in a terminal state — while a Codex review runs concurrently against the local diff. Then fetches unresolved threads from all three code-review sources plus failed CI checks, validates each against the actual code / log, handles invalid ones in place (reply on thread / NOSONAR comment), and presents a unified plan for fixing the valid ones (split into logical groups if more than 3).
Manage the Cowork service and Claude Desktop. Use when the user wants to start, stop, or reinstall cowork.
Use when the user asks to transcribe a YouTube video, get a YouTube transcript, or download/save a transcript from a YouTube URL.
Use when the user asks to draft release notes, write release notes, or summarize changes since the last release for the current repository.
| name | make-skill |
| description | Guide for creating valid Claude Code skills. Use when user asks how to create a skill, make a skill, or needs help with skill structure. |
This skill documents how to create valid Claude Code skills.
Skills MUST be in a .claude/skills/ folder - Claude Code only scans these locations:
~/.claude/skills/your-skill-name/SKILL.md<project-root>/.claude/skills/your-skill-name/SKILL.mdSkills placed anywhere else will NOT be discovered.
SKILL.md (case-sensitive - not skill.md or SKILL.MD)notion-project-setup)your-skill-name/ # kebab-case folder name
├── SKILL.md # REQUIRED - exact filename
├── scripts/ # Optional - executable code
├── references/ # Optional - documentation
└── assets/ # Optional - templates, examples
Every SKILL.md must start with YAML frontmatter:
---
name: your-skill-name
description: When Claude should invoke this skill
---
The description field is critical - Claude uses it to decide when to automatically invoke the skill.
Add these to frontmatter as needed:
user-invocable: false - Only Claude can invoke automatically (user cannot use /command)disable-model-invocation: true - Only user can invoke via /command (Claude cannot auto-invoke)context: fork - Run with specific agent type---
name: my-task
description: Use when user asks to perform my-task or mentions my-task workflow
---
# My Task Skill
Instructions for Claude when this skill is invoked...
## Steps
1. First do X
2. Then do Y
3. Finally do Z
---
name: deploy
description: Deploy the application to production
disable-model-invocation: true
---
User invokes with /deploy - Claude cannot auto-invoke.
SKILL.md (case-sensitive)~/.claude/skills/ or <project>/.claude/skills/name and description