The Agent Skills Directory

[SAFE]: The skill consists entirely of Markdown documentation, templates, and reference materials intended to guide users and agents in creating portable skill packages. It does not include any executable scripts, binary files, or configuration settings that perform automated actions.- [COMMAND_EXECUTION]: The workflow section describes the use of the agentskills validation tool via the command uvx --from skills-ref agentskills validate. This is a standard development practice for linting and verifying compliance with the Agent Skills specification and does not involve arbitrary or malicious command execution.- [EXTERNAL_DOWNLOADS]: The documentation references the skills-ref package available on PyPI. This is an official reference implementation for the specification described in the skill and is used solely for validation purposes within the described workflow.- [NO_CODE]: There are no scripts located in the scripts/ directory; all content is descriptive or template-based, reducing the attack surface to zero for runtime execution vulnerabilities.

authoring-skills