The Agent Skills Directory

[SAFE]: The skill functions as a navigational aid and contains no executable scripts or dangerous commands. It is composed entirely of markdown documentation and YAML configuration.
[SAFE]: No network exfiltration, sensitive file access, or privilege escalation patterns were found. The skill does not perform any operations involving external data or remote scripts.
[SAFE]: The instructions are descriptive and intended for human guidance or agent routing. There are no attempts to override safety filters, obfuscate content, or establish persistence.
[SAFE]: The metadata in agents/openai.yaml correctly limits the skill's interface and disables implicit invocation, adhering to best practices for non-tool-invoking skills.

ask-matt