finishing-a-development-branch
Pass
Audited by Gen Agent Trust Hub on Jul 20, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing data from the local repository state and user input that is subsequently used in executable shell commands.
- Ingestion points: Git branch names retrieved via git rev-parse and git merge-base, as well as branch names and menu choices provided by the user at runtime.
- Boundary markers: No delimiters or safety instructions are used to separate untrusted data strings from the shell command logic.
- Capability inventory: The skill possesses the capability to execute arbitrary commands through PowerShell via npm test, git push, git checkout, and git merge operations.
- Sanitization: There are no instructions provided to sanitize, validate, or properly quote inputs before they are passed to the shell.
- [COMMAND_EXECUTION]: The skill executes various PowerShell commands incorporating variables like and . If these variables are populated with malicious strings from the repository environment or user input (e.g., using shell metacharacters), it could lead to unintended command execution.
Audit Metadata