git-cleanup
Pass
Audited by Gen Agent Trust Hub on Apr 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses Git CLI commands to survey and remove local branches and remotes. It includes logic to safeguard the 'main' branch and the currently active branch from deletion.
- [PROMPT_INJECTION]: There is a potential for indirect prompt injection because the skill reads and processes local Git branch and remote names, which are external data sources.
- Ingestion points: Data provided by
git branchandgit remotein SKILL.md. - Boundary markers: None used to encapsulate the output of the survey commands.
- Capability inventory: Capabilities include branch and remote removal via the Git CLI as defined in SKILL.md.
- Sanitization: The skill filters for the 'main' branch using grep but does not verify the content of other branch names for malicious instructions before presentation to the agent.
Audit Metadata