Skill Creator
If you're building skills for Claude, this is the meta tool you want. It walks you through the full cycle: drafting the skill, writing test prompts, running evals in the background while helping you set up quantitative benchmarks, then iterating based on results. The progressive disclosure guidance is solid (metadata always loads, skill body loads on trigger, bundled resources load as needed), and there's a separate flow for optimizing skill descriptions to fix undertriggering. What I like is the flexibility: it meets you wherever you are in the process, whether you're starting from scratch or refining an existing draft. Just be aware it assumes familiarity with the skill format and evaluation workflows.
- skills.sh installs
- 208
- GitHub stars
- 1.7K
- last updated
- Apr 25, 2026
Install
$ npx -y skills add feiskyer/claude-code-settings --skill skill-creator --agent claude-codeInstalls 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 the agent with the skill loaded 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.pyscript 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
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageDetails
- Repository
- github.com/feiskyer/claude-code-settings
- Categories
- Code Review & Quality, AI & Agent Building, Skill Development
- License
- MIT
- First seen
- Jun 11, 2026
More from feiskyer/claude-code-settings
All 15Recommended
More Code Review & Quality
complianceanthropics/knowledge-work-pluginsNavigate privacy regulations (GDPR, CCPA), review DPAs, and handle data subject requests. Use when reviewing data processing agreements, responding to data subj207skills.sh installs
docs-updatewarpdotdev/oz-skillsUpdate user-facing documentation when code changes. Use when asked to update docs, review docs, handle documentation changes, run scheduled documentation tasks,206skills.sh installs
implementyonatangross/orchestkitFull-power feature implementation using parallel subagents for backend, frontend, testing, and security, with worktree isolation and quality verification in one206skills.sh installs
edge-ttsvaibhav0806/trying-somethingText-to-speech conversion using `uvx edge-tts` for generating audio from text. Use when (1) User requests audio/voice output with the "tts" trigger or keyword. 203skills.sh installs
semantic-gitsiviter-xyz/dot-agentManage Git commits using conventional commit format with atomic staging. Always generate plain git commands before running them and offer to let the user run th201skills.sh installs
input-validation-xss-preventionharperaa/secure-claude-skillsValidate and sanitize user input to prevent XSS, injection attacks, and ensure data quality. Use this skill when you need to validate forms, sanitize user input197skills.sh installs