The Agent Skills Directory

[SAFE]: No security threats or malicious patterns were detected. The skill is composed of purely instructional Markdown files and metadata configurations. It does not include any scripts, executables, or network-enabled tools.- [PROMPT_INJECTION]: The skill contains well-defined operational boundaries, principles, and 'Ask First' protocols that guide agent behavior and prevent instructions from being overridden by untrusted inputs.- [DATA_EXFILTRATION]: The skill does not access sensitive local files (such as SSH keys or environment variables) and does not perform any network operations to external domains. Communication is logically restricted to platform-defined handoffs between agents.- [REMOTE_CODE_EXECUTION]: There are no instructions for package installation, remote script downloading (e.g., via curl or wget), or dynamic code execution.- [INDIRECT_PROMPT_INJECTION]: While the skill processes external data from users and other agents, it does not possess high-privilege capabilities or tools that would allow for exploitation. Furthermore, its 'Grounding' principle requires explicit evidence for claims, which serves as a defensive measure against processing misleading instructions embedded in data.

pmm