The Agent Skills Directory

[SAFE]: The skill instructions are limited to defining a process for creating markdown files within a local docs/ directory. No command execution, file system access outside the project scope, or network interactions are defined.
[INDIRECT_PROMPT_INJECTION]: The skill operates on user-provided data from brainstorming or design sessions to generate documentation. While it does not explicitly define boundary markers or sanitization, the risk is minimal as the skill lacks high-privilege capabilities such as code execution or external network exfiltration.

design-doc