The Agent Skills Directory

[SAFE] (SAFE): No malicious behavior or patterns detected. The skill is a legitimate utility for managing other skills via user-directed updates.\n- [Indirect Prompt Injection] (LOW): The skill provides a surface for modifying agent instructions through its editing capability.\n
Ingestion points: Skill files located in the .claude/skills/ directory.\n
Boundary markers: None present in the instructions to distinguish between administrative logic and updated content.\n
Capability inventory: Read, Edit, and Glob tools allow filesystem modification within the workspace.\n
Sanitization: No content sanitization is performed, though basic YAML structure is validated during the update process.

update-skill