Summary
If you're building skills for Claude or other agents and tired of validation errors, missing metadata, or bloated instructions, this handles the scaffolding. It spins up new skills with the right folder structure, generates agent YAML files from explicit parameters, validates frontmatter and dependencies, and packages everything into a distributable zip. The workflow pushes you toward focused, triggerable skills with short SKILL.md files and detailed references tucked away. It's opinionated about keeping things minimal and mechanically correct, which is exactly what you want when you're maintaining more than a couple of these.
Install to Claude Code
npx -y skills add dkyazzentwatwa/chatgpt-skills --skill skill-creator --agent claude-codeInstalls into .claude/skills of the current project.
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →
Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →
Files
SKILL.mdView on GitHub
Skill Creator
Design or refactor skills so they are small, triggerable, and mechanically valid.
Workflow
- Define the exact user requests that should trigger the skill.
- Keep
SKILL.mdshort and put deep detail inreferences/. - Add only the scripts, assets, or references that make execution materially more reliable.
- Generate
agents/openai.yamland validate before packaging.
Scripts
scripts/init_skill.pyinitializes a new skill with optional resource folders andagents/openai.yaml.scripts/generate_openai_yaml.pywritesagents/openai.yamlfrom explicit interface values.scripts/quick_validate.pyenforces strict frontmatter, directory naming, agent metadata, and script dependency rules.scripts/package_skill.pyvalidates first, then zips the skill for distribution.
Rules
- Keep frontmatter to
nameanddescriptiononly. - Make
descriptiondo the trigger work: what the skill does and when to use it. - Match the folder name exactly to
name. - Prefer one focused workflow over a kitchen-sink skill.
- Remove placeholder files and generic boilerplate before publishing.
Typical Commands
python3 skill-creator/scripts/init_skill.py my-skill --path .
python3 skill-creator/scripts/generate_openai_yaml.py my-skill \
--interface display_name="My Skill" \
--interface short_description="One-line summary" \
--interface default_prompt="Help me use this skill."
python3 skill-creator/scripts/quick_validate.py my-skill
python3 skill-creator/scripts/package_skill.py my-skill ./dist
Featured
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →First SeenJun 3, 2026
