Create and Evaluate Skills
Turn manual workflows into agent skills, improve existing skills, or evaluate their quality. Your AI guides you through skill creation, benchmarking, and targeted improvements using best-practice patterns.
Try asking: “Create a skill to summarize daily Slack messages”
himself65 on GitHub
Curated by PromptsEdge from a public repo · MIT license. All credit goes to the author.
What is this skill?
Building high-quality AI agent skills can be challenging, especially when you want to automate repeated workflows or ensure your skills meet the latest standards. With this tool, your AI can help you create new skills from scratch, improve existing ones, or evaluate their quality using a detailed scoring rubric. This is perfect for anyone developing custom skills for Claude, ChatGPT, or similar assistants—whether you're a developer, product manager, or power user looking to automate and optimize your processes.
What you can do
- Create a new skill with guided architecture planning, SKILL.md writing, and reference file setup ("Build a skill for exporting Slack messages").
- Improve an existing skill by scoring it and applying ranked suggestions ("Optimize this skill for reliability").
- Evaluate a skill's quality using a 10-dimension rubric ("Score this skill for clarity and output contract").
- Benchmark or run evals on a skill to identify strengths and weaknesses.
- Turn a manual, repeated workflow into a reusable skill ("Turn my weekly report process into a skill").
How it works
When you ask your AI to create, improve, or evaluate a skill, it walks you through structured prompts: from defining the skill's purpose and triggers, to drafting SKILL.md, to setting up reference files and output contracts. For evaluation, it scores your skill across ten quality dimensions, referencing best-practice patterns and suggesting targeted improvements. The AI adapts its approach based on your environment—choosing the right detection flows and fallback methods for any external tools involved.
Good to know
- Skill file creation requires a CLI platform (Claude Code or similar); evaluation and planning work on Claude.ai.
- No external accounts or API keys are needed unless your skill interacts with third-party services.
- Reference files provide detailed examples, scoring rubrics, and writing guidance for best results.
What’s inside
Install this skill
- 1
Get the skill — it’s free
Use the Get this skill panel. Unlocked skills stay in My skills. - 2
Download or clone the files
Download the zip, or clone the repo and copy theplugins/skill-creator/skills/skill-creatorfolder. - 3
Put it where your agent looks for skills
For Claude Code, use your personal skills folder (every project) or a project’s own folder:~/.claude/skills/skill-creator/SKILL.md # all projects .claude/skills/skill-creator/SKILL.md # this project only
- 4
Just ask
No command needed. The agent reads the skill’s description and loads it on its own when your request matches.
SKILL.md frontmatter
What your agent reads to decide when to load this skill.
---
name: skill-creator
description: Create, improve, and evaluate agent skills (SKILL.md plus reference files). Use this skill whenever the user wants to build, scaffold, or design a new skill, improve or fix an existing skill that isn't working well, score or benchmark a skill's quality or run evals on it, or turn a repeated manual workflow into a skill ("I keep doing X manually", "can you remember how to do X", "turn this into a skill").
---Files
Open any Markdown file to read it here.
Related skills
View all →- Create and Improve SkillsFreeYour AI helps you create new skills or improve existing ones by drafting, testing, evaluating, and rewriting them. It runs test prompts, analyzes results quantitatively and qualitatively, and guides you through iterative refinement.
- Audit Workspace CapabilitiesFreeYour AI audits your repo, servers, plugins, and environment to reveal current capabilities and recommend next steps. It identifies available skills, hooks, agents, and missing integrations without exposing secrets.
- Verify Claude Code SessionsFreeYour AI runs a multi-step verification for Claude Code projects before completion. It checks builds, types, linting, tests, security scans, and code diffs to ensure quality and readiness for pull requests.
- Share Durable Agent MemoryFreeYour AI shares and resumes work state across different agents using the ECC Memory Vault. It stores and retrieves portable Markdown context files to hand off tasks and search project knowledge across sessions.