The Agent Skills Directory

[SAFE]: No security threats were identified. The skill defines a set of validation rules and provides instructions for the agent to check repository contents for compliance.
[COMMAND_EXECUTION]: The skill suggests using the fd utility to identify directories for validation. This is a standard filesystem discovery operation with no privilege escalation or dangerous flags.
[INDIRECT_PROMPT_INJECTION]: The skill involves reading and reporting on content from SKILL.md files in the repository. While this creates a point of ingestion for external data, the skill does not grant the agent dangerous capabilities or provide paths for data exfiltration based on that content.
Ingestion points: SKILL.md files in subdirectories of skills/.
Boundary markers: None specified.
Capability inventory: File reading and text output only; no subprocess execution, network tools, or file-write capabilities are requested.
Sanitization: None requested.

validate-skills