Skill Composer

Use when the user asks to create a new Agent Skill, revise an existing skill's authored behavior or activation, audit a complete skill package, design its eval suite, port it across agent harnesses, or prepare and validate its release. When a generic skill-authoring helper overlaps, this skill owns scope, structure, portability, validation, and release history. If the work neither changes nor audits an Agent Skill package, do not load this skill.

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add bolasblack/claude-skills --skill skill-composer
Manual install
git clone --depth 1 https://github.com/bolasblack/claude-skills.git /tmp/claude-skills
mkdir -p ~/.claude/skills
cp -r /tmp/claude-skills/skills/skill-composer ~/.claude/skills/skill-composer

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Skill Composer

When modifying, reviewing, testing, or releasing Skill Composer itself, first read [SPEC.md](SPEC.md) in full. Otherwise do not inspect its eval manifests, fixtures, package tests, or eval runner implementation; Step 7 says when to run its scripts' public CLIs, and the target needs no runner copy.

## Authoring Authority

Skill Composer owns skill-authoring policy. Use an injected helper such as `skill-creator` only for platform scaffolding, schema validation, or evaluation; its authoring advice is advisory, but a target's verified format requirement binds.

## Core Design Principles

### Bounded Entry Point

These rules apply to every skill you create, update, or review:

- `SKILL.md` stays within **15,000 bytes** ([why](SPEC.md)).
- Keep every standing rule (one that applies across a whole run) in `SKILL.md`, must-follow gates first. A reference holds one branch's workflow or lookup material.
- Point at whole reference files, each with its read condition, and tell the agent to read the file in full; a file tool cannot jump to an anchor, so a section pointer gets a fragment read.
- Point to another skill by telling the agent to **invoke it by name through the host's skill mechanism** (in Claude Code, the Skill tool). **Reading its `SKILL.md` with a file tool is not invocation**: an agent reading the file tends to take only part of it and miss gates written elsewhere ([why](SPEC.md)).
Read the full skill on GitHub

Details

Skill name
skill-composer
Path
skills/skill-composer/SKILL.md
Repo updated
18 November 2026

FAQ

What does the Skill Composer skill do?

Use when the user asks to create a new Agent Skill, revise an existing skill's authored behavior or activation, audit a complete skill package, design its eval suite, port it across agent harnesses, or prepare and validate its release. When a generic skill-authoring helper overlaps, this skill owns scope, structure, portability, validation, and release history. If the work neither changes nor audits an Agent Skill package, do not load this skill.

How do I install the Skill Composer skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add bolasblack/claude-skills --skill skill-composer. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Skill Composer popular?

It lives in bolasblack/claude-skills, which has 5 GitHub stars and 0 forks.

Related skills

Caveman

5

bolasblack/claude-skills

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

Images, Video & Audio

Agent Centric

5

bolasblack/claude-skills

Track decisions (AGD) with validation and indexing. Use when making design decisions, recording important choices, discussing trade-offs, or when user mentions AGD or decision record.

AI, LLMs & Agents

Codex

5

bolasblack/claude-skills

OpenAI Codex second opinion via the codex CLI — three modes. Review: independent diff review with a pass/fail gate. Challenge: adversarial pass that tries to break the code. Consult: ask Codex anything, with session continuity for follow-ups. Use when the user says "codex review", "codex challenge", "ask codex", or wants a second opinion from Codex.

Software Development

Skilzy Creator

8

Freddy-CHEN-ux/00_claude_skills

Comprehensive toolset for creating, converting, validating, and packaging Skilzy-compliant AI agent skills. This skill should be used when users want to initialize a new skill, convert a Claude skill archive to the Skilzy format, validate an existing skill's structure and compliance, or package a skill for distribution to the Skilzy registry.

AI, LLMs & Agents

Router Engineering

8

Freddy-CHEN-ux/00_claude_skills

Master orchestration for routing technical problems through 32 engineering skills - AI/ML, software development, data, APIs, and Claude Code framework

AI, LLMs & Agents

Statechart

5

bolasblack/claude-skills

Check a TLA+ model of a system that moves between states, such as a product flow, a protocol, or a service, and draw it as a Mermaid statechart. The report gives a shortest path for each rule that breaks. Use for /statechart, or when a task needs a TLA+ model checked or a model's bug shown as a path through states.

AI, LLMs & Agents

Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.