A meta-skill for building and refining other skills through structured iteration. Walk through the full cycle: draft the skill, write test prompts, run evaluations both qualitative and quantitative, then rewrite based on results. It handles users at any entry point, whether they're starting from scratch or already have a draft to optimize. The eval-viewer scripts give you concrete metrics to improve against, and there's a separate flow for tuning skill descriptions to fix undertriggering. Worth noting the guidance on progressive disclosure: keep the main SKILL.md under 500 lines and push reference material into bundled files. Most useful when you're serious about iteration rather than one-shot skill writing.

skills.sh installs
8.3K
GitHub stars
46.7K
last updated
Aug 21, 2026

Install

$ npx -y skills add anthropics/claude-plugins-official --skill skill-creator --agent claude-code

Installs into .claude/skills of the current project.

Files

Skill Creator

A skill for creating new skills and iteratively improving them.

At a high level, the process of creating a skill goes like this:

  • Decide what you want the skill to do and roughly how it should do it
  • Write a draft of the skill
  • Create a few test prompts and run claude-with-access-to-the-skill on them
  • Help the user evaluate the results both qualitatively and quantitatively
    • While the runs happen in the background, draft some quantitative evals if there aren't any (if there are some, you can either use as is or modify if you feel something needs to change about them). Then explain them to the user (or if they already existed, explain the ones that already exist)
    • Use the eval-viewer/generate_review.py script to show the user the results for them to look at, and also let them look at the quantitative metrics
  • Rewrite the skill based on feedback from the user's evaluation of the results (and also if there are any glaring flaws that become apparent from the quantitative benchmarks)
  • Repeat until you're satisfied
  • Expand the test set and try again at larger scale

Sponsors

Details

Repository
github.com/anthropics/claude-plugins-official
Categories
Testing & QA, Skill Development, Debugging
First seen
May 16, 2026

More from anthropics/claude-plugins-official

All 24
1onmax avatarvueonmax/nuxt-skillsUse when editing .vue files, creating Vue 3 components, writing composables, or testing Vue code - provides Composition API patterns, props/emits best practices8.1Kskills.sh installs
2rivet-dev avatarrivetkitrivet-dev/skillsRivetKit backend and Rivet Actor runtime guidance. Use for building, modifying, debugging, or testing Rivet Actors, registries, serverless/runner modes, deploym7.7Kskills.sh installs
3rorkai avatarasc-testflight-orchestrationrorkai/app-store-connect-cli-skillsOrchestrate TestFlight distribution, groups, testers, and What to Test notes using asc. Use when rolling out betas.7.6Kskills.sh installs
4trailofbits avatarcode-maturity-assessortrailofbits/skillsSystematic code maturity assessment using Trail of Bits' 9-category framework. Analyzes codebase for arithmetic safety, auditing practices, access controls, com7.5Kskills.sh installs
5google avatargoogle-agents-cli-onboardinggoogle/skillsOnboarding entrypoint for agents-cli in Agent Platform. It should be used when the user wants to "create a new agent", "develop an agent", "build an agent using7.3Kskills.sh installs
6wondelai avatarjobs-to-be-donewondelai/skillsDiscover what customers truly need by analyzing the "job" they hire your product to do. Use when the user mentions "customer discovery", "why customers churn", 7.3Kskills.sh installs