The Agent Skills Directory

[NO_CODE]: This skill is entirely comprised of Markdown files containing text-based instructions and reference materials. There are no scripts (e.g., .sh, .py, .js) or binary executables included.
[SAFE]: The skill was analyzed for all 11 threat categories, including prompt injection, data exfiltration, and obfuscation. No malicious patterns or security vulnerabilities were found. The skill's instructions to read project-specific context files (.agents/product-marketing.md) are standard practices for tailoring agent responses and do not involve sensitive system paths.

offers