#519 · Primary category: AI Agents & Automation

agent-skill-creator

agent-skills ai-agents claude-code codex cursor developer-tools evals github-copilot governance skill-marketplace

Build tested agent skills and govern their lifecycle through a user-defined marketplace: evidence, discovery, updates, rollback, quarantine, and 17-platform distribution.

Project last updated:08/28/26

GitHub Stars

2.4K

Forks

258

Contributors

3

License

MIT

Why we included this project

Teams that keep re-explaining the same workflows to every coding assistant will get real mileage out of this. You hand it a plain-English description of a task you repeat, say a weekly CRM cleanup that ends in a PDF report or a stock analysis, and it drafts a packaged skill with working code and its own evaluation spec, then runs a security scan and installs the result across Claude Code, Cursor, Copilot, Windsurf, and a dozen other assistants with one command. A built-in pipeline validates the skill before anything is installed, and an evolve loop feeds real failure evidence back into regeneration, so the skill stays reliable after you stop watching it. The practical payoff is cross-platform consistency: one authoring pass replaces maintaining separate prompt files and spec formats for every tool your team uses.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category