The Agent Skills Directory

[SAFE]: The skill performs local file system operations using Grep and Read strictly to discover and document other skills within the src/skills directory. No sensitive files or network resources are accessed during these operations.
[SAFE]: The skill's ingestion of local skill metadata (names and descriptions) constitutes a standard documentation task. While the agent processes content from these files, the limited scope of the extraction and the absence of dangerous tools (such as shell execution or network requests) mitigates the risk of indirect prompt injection.

help