The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection because it reads and processes data from external files without sanitization.
Ingestion points: The skill uses the Read tool to ingest content from a DESIGN.md file located within the user's project (SKILL.md).
Boundary markers: Ingested content is formatted under a "DESIGN SYSTEM (REQUIRED)" header, which provides structural separation but does not strictly prevent the underlying model from following instructions embedded within the file (SKILL.md).
Capability inventory: The skill is granted access to Read and Write tools, allowing it to interact with the local file system (SKILL.md).
Sanitization: There is no evidence of content filtering or instruction scrubbing for the data interpolated from DESIGN.md before it is included in the final enhanced prompt.
[SAFE]: The skill references official Google documentation for prompting best practices, which is a trusted service (SKILL.md).
[SAFE]: The installation instructions utilize a standard package manager and point to a repository associated with a known organization (README.md).

enhance-prompt