The Agent Skills Directory

[SAFE]: No security issues were detected. The skill is purely instructional and designed for requirements gathering and documentation.
[COMMAND_EXECUTION]: The skill manages markdown files within a specific path (workflows/*.md and NOTES.md) in the local workspace. These are legitimate file operations for its stated purpose of managing workflow specifications.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from NOTES.md, which could potentially harbor malicious instructions if a user populates it with untrusted content. However, because the skill lacks dangerous capabilities such as network access, system privilege escalation, or arbitrary code execution, this represents a negligible risk surface.

loop-me