The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional markdown without any embedded scripts, binary files, or network-enabled commands.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to process and execute tasks from an external plan file, which creates an indirect prompt injection surface. The ingestion point is the plan file referenced in SKILL.md. Boundary markers are not explicitly provided. The capability inventory includes potential tool usage for implementing steps and running verifications as described in the plan. The skill includes sanitization via a mandatory 'review critically' step to identify concerns before execution and uses human-in-the-loop checkpoints for verification output and feedback.

executing-plans