The Agent Skills Directory

[SAFE] (SAFE): No security issues detected. The skill is purely instructional and focuses on methodology for task breakdown and verification.
[DATA_EXFILTRATION] (INFO): The skill instructs the agent to save planning files in the project root using dynamic names derived from tasks ({task-slug}.md). While this could lead to root directory clutter, it does not constitute a security risk provided the agent's file-writing capabilities (not part of this skill's scope) handle path sanitization.
[COMMAND_EXECUTION] (INFO): The skill provides examples of commands like npx create-next-app or curl as part of its 'Verification' principles. These are examples for documentation purposes and are not executed by the skill itself, which is restricted to Read, Glob, and Grep tools.

plan-writing