batch
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
Bashtool to perform git operations and execute tests, which is consistent with its stated purpose. - [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection.
- Ingestion points: External instructions enter the agent context via the
$ARGUMENTSparameter. - Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands within the prompt logic.
- Capability inventory: The orchestration grants sub-agents access to powerful tools like
Bash,Write, andEditto perform filesystem modifications. - Sanitization: The skill does not validate or sanitize the provided instructions before propagating them.
- [DYNAMIC_EXECUTION]: The skill dynamically creates task units and spawns new agent instances using the
Agenttool based on the user's refactoring request.
Audit Metadata