The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The README describes an installation process that retrieves the skill from a Google-hosted repository.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from an external DESIGN.md file without adequate sanitization or boundary enforcement, creating a surface for potential instruction injection.
Ingestion points: SKILL.md (Step 2) reads the project's DESIGN.md file.
Boundary markers: None; the skill lacks specific markers to separate untrusted file content from its own instructions.
Capability inventory: SKILL.md utilizes Read and Write tools, including instructions to write to local files like next-prompt.md in the Output Options section.
Sanitization: The skill does not perform validation or filtering on the content extracted from DESIGN.md.

enhance-prompt