The Agent Skills Directory

[SAFE]: The skill functions as a documentation and methodology guide for creating other agent skills. It promotes security-positive practices such as robust testing and deterministic process enforcement.
[SAFE]: Frontmatter configuration uses disable-model-invocation: true, which appropriately limits the skill to manual user invocation, reducing the risk of unintended automated triggers.
[SAFE]: References to internal utility scripts (scripts/lint-skill-evals.py, scripts/lint-skill-frontmatter.py) are for local linting and validation of skill structure and do not involve remote code execution or external downloads.
[SAFE]: The skill explicitly instructs on identifying and countering agent 'rationalizations'—a technique that improves agent alignment and resistance to instruction bypass.

author-skills