The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest external specifications or requirements to generate implementable tasks, creating a potential surface for indirect prompt injection.
Ingestion points: Processes task requirements and specifications provided to the agent in SKILL.md.
Boundary markers: The instructions lack specific delimiters or warnings to ignore malicious instructions that might be embedded within the input requirements.
Capability inventory: The skill includes file writing capabilities (to the docs/plans/ directory) and initiates code execution workflows through sub-skills such as superpowers:executing-plans and superpowers:subagent-driven-development.
Sanitization: No input validation or sanitization is specified for the external requirements before they are incorporated into implementation plans.

writing-plans