plugin official · Anthropic

skill-creator — Claude Code plugin: setup and skill evals

Q: How do I install skill-creator in Claude Code?

Add the marketplace with `/plugin marketplace add anthropics/claude-plugins-official`, then install the plugin with `/plugin install skill-creator@claude-plugins-official`.

Create or improve skills, run evals, and measure performance with variance analysis. Use it when a skill needs testing before adoption.

Install into Claude Code · Plugin marketplace All methods and platforms ↓

/plugin install skill-creator@claude-plugins-official
/plugin marketplace add anthropics/claude-plugins-official

Open on GitHub

What skill-creator does

The plugin creates new skills from scratch and improves existing ones. It can also run evals to test how a skill performs and measure its results.

For deeper comparison, skill-creator benchmarks performance with variance analysis. That makes it possible to examine consistency across evaluations instead of relying on a single run.

This plugin is built for Claude Code, so testing it on another platform is not a reasonable expectation.

What it helps with

create a skill from scratch
improve an existing skill
run evals for a skill
measure skill performance
benchmark results with variance analysis

Example prompts for the agent

Create a new skill from scratch for this task and suggest its structure.
Improve the existing skill and assess how its performance changes.
Run evals for this skill and benchmark the results with variance analysis.

How to install skill-creator

Plugin marketplace

/plugin install skill-creator@claude-plugins-official

/plugin marketplace add anthropics/claude-plugins-official

Step by step, with commands for each platform: Claude Code plugins: marketplaces, installing, and building your own

Questions about skill-creator

How do I install skill-creator in Claude Code?

Add the marketplace with /plugin marketplace add anthropics/claude-plugins-official, then install the plugin with /plugin install skill-creator@claude-plugins-official.

What does skill-creator do?

It creates new skills, improves existing ones, runs evals, and measures skill performance with variance analysis.

Can skill-creator compare skill results?

Yes. The plugin benchmarks skill performance and includes variance analysis for the results.

Source text

skill-creator · Apache-2.0

# skill-creator

Create new skills, improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, update or optimize an existing skill, run evals to test a skill, or benchmark skill performance with variance analysis.

bmad-eval bmad-code-org Runs automated benchmarks and evaluations for agent skills across four modes, checking trigger accuracy and grading quality in a clean workspace. CX Codex +1 4,600installs caveman-manage JuliusBrussee Checks Caveman Cloud experiment state and evidence, then recommends one action or blocks an unsafe request. Use it before starting, approving, cancelling, or rolling back an experiment. CX Codex +1 104,200installs deepeval Confident AI DeepEval skill sets up evaluation loops for AI agents and RAG systems. It creates pytest suites, generates golden datasets, and catches regressions. CX Codex +1 2,800installs agent-eval colbymchenry Compares agent behavior with and without CodeGraph on a real repository. Use it to audit retrieval quality and test a local or published CodeGraph version. CX Codex +1 1,800installs arize-experiment official arize Runs Arize experiments with the ax CLI to benchmark models and export evaluation scores across dataset versions. 1,100installs ponytail DietrichGebert ponytail steers Claude Code toward small, direct solutions and away from abstractions nobody asked for. Use it when you want the agent to build on what the project already has. ★ 176,443repo stars impeccable pbakaus Hands your agent a designer's vocabulary: 24 commands like polish, audit and bolder, so UI feedback stops being “make it pretty”. ★ 88,212repo stars last-30-days-skill mvanhorn Searches the web for what people posted about a topic in the last month and turns it into one structured brief. ★ 65,508repo stars

From the same set123

The whole set →