The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the bash tool to perform file system operations relevant to skill management. This includes listing files with ls, creating directory structures with mkdir -p, removing directories with rm -r, and managing version control with git checkout. These commands are standard for a developer utility and are scoped to the user's local skill directories.
[SAFE]: The skill exposes an indirect prompt injection surface due to its core functionality of reading and auditing external skill files.
Ingestion points: Content from existing skills is ingested into the agent's context in workflows/audit-skill.md, workflows/edit-skill.md, and workflows/upgrade-to-router.md.
Boundary markers: No specific delimiters or instructions to ignore embedded commands are present in the workflows.
Capability inventory: The skill is compatible with and utilizes bash, read, write, and edit tools.
Sanitization: There is no evidence of sanitization or content validation for the ingested skill data.
Assessment: This surface is an inherent part of the skill's primary purpose as an auditing tool and does not indicate malicious intent. No evidence of exploitation was found in the provided files.

author-skill