The Agent Skills Directory

[SAFE]: The skill serves as a template for local plugin development and does not contain any executable code or remote dependencies.
[SAFE]: Security best practices are explicitly integrated into the workflow, including instructions to avoid hardcoding secrets and to use environment-backed headers/tokens.
[SAFE]: The lifecycle instructions require mandatory human-in-the-loop verification steps (commands like /plugin trust and /plugin enable), ensuring that new plugins are reviewed before activation.

plugin-creator