The Agent Skills Directory

[SAFE]: The skill is purely instructional, focusing on project documentation structure (README, MkDocs, application requirements). It does not contain executable scripts, subprocess calls, or network requests.
[DATA_EXFILTRATION]: The skill explicitly warns against data exposure, directing users and agents not to include secrets, private keys, or PII in files like AGENTS.md or logs. It defines clear boundaries for sensitive information.
[SECURITY_BEST_PRACTICES]: The documentation references (architecture-and-cross-cutting.md) provide proactive security guidance, such as enforcing server-side authorization and distinguishing between authentication and ownership.
[REMOTE_CODE_EXECUTION]: No remote code patterns or package installation commands were detected. References to tools like MkDocs, Mermaid, and D2 are standard for documentation workflows.

project-documentation