The Agent Skills Directory

[SAFE]: The skill operates as a structural organization tool for agent workflows. All logic is contained within standard markdown files and focuses on methodology rather than execution of dangerous commands.
[INDIRECT_PROMPT_INJECTION]: The skill includes instructions to read and write repository files (in references/review-skill-router/SKILL.md and references/implement-skill-router/SKILL.md). While this creates an ingestion surface for potentially untrusted data if used on third-party repositories, the primary purpose is for development and management of the agent's own capabilities. No malicious injection triggers were found.
[COMMAND_EXECUTION]: While the skill discusses implementation, the actual commands are limited to local file system operations (create/update SKILL.md) within a specified target directory, which is consistent with the skill's stated purpose.

skill-routers