skills/shunsukehayashi/miyabi-claude-plugins/Git Workflow with Conventional Commits/Gen Agent Trust Hub
Git Workflow with Conventional Commits
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION] (SAFE): Utilizes standard development tools including
git,gh(GitHub CLI), andcargofor version control, PR management, and code quality checks. All commands are consistent with the skill's primary purpose. - [DATA_EXFILTRATION] (SAFE): While the skill pushes data to remote repositories (via
git pushandgh pr create), these actions are intentional and core to a Git workflow. No patterns for exfiltrating sensitive system files (e.g., SSH keys, environment variables) were found. - [PROMPT_INJECTION] (SAFE): No instructions designed to override agent safety filters, extract system prompts, or bypass constraints were identified. The instructional language is natural and focused on task execution.
- [EXTERNAL_DOWNLOADS] (SAFE): The skill mentions external tools like
git-filter-repofor remediation but does not include automated remote script execution or unverifiable package installations.
Audit Metadata