update-base-branch
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the Git repository environment, such as branch names and configuration, which could influence the agent's behavior.
- Ingestion points: Git repository metadata retrieved via CLI commands (SKILL.md).
- Capability inventory: Execution of
gitshell commands includingfetch,switch, andmerge(SKILL.md). - Boundary markers: The instructions explicitly require the agent to stop and ask the user for confirmation if targets or remotes are ambiguous (SKILL.md).
- Sanitization: The skill enforces strict checks, such as worktree cleanliness and fast-forward-only requirements, to prevent unintended changes to the repository state (SKILL.md).
- [COMMAND_EXECUTION]: The skill functions by executing shell commands using the
gitCLI tool. This is the intended primary purpose of the skill and is documented as a safe, non-destructive operation.
Audit Metadata