The Agent Skills Directory

[SAFE]: The skill establishes formatting rules for Git commit messages, including ticket prefixes and keyword types, which is an informational and procedural task without security implications.
[SAFE]: Usage of local environment variables like ${CLAUDE_PLUGIN_ROOT} and references to local documentation (e.g., CLAUDE.md) are standard for development-focused skills and do not involve unauthorized file access or network activity.
[SAFE]: The instructions for the agent to trigger internal quality gate checks (like the perform-preflight skill) align with secure development practices and do not involve untrusted remote code execution.

committing-changes