daily
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill implements a hook mechanism in
commands/start.mdandcommands/end.mdthat reads and executes instructions from.groove/hooks/start.mdand.groove/hooks/end.md. This constitutes an indirect prompt injection surface. - Ingestion points: Hook files located at
.groove/hooks/start.mdand.groove/hooks/end.md. - Boundary markers: No explicit delimiters or safety instructions are used when parsing the content of these files.
- Capability inventory: The agent utilizes
Read,Write,Edit, andBash(git:*)tools to carry out hook instructions. - Sanitization: The skill does not validate or sanitize the list of actions retrieved from the hook files before execution.
- [COMMAND_EXECUTION]: The skill uses the Bash tool to execute
git rev-parse --show-toplevel. While restricted to the git binary, this is an execution of an external process to determine environment context.
Audit Metadata