test-data-builder
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill requests access to
Bash(cmd:*), which allows for the execution of arbitrary shell commands. This is a high-privilege capability within the agent's environment, although it is consistent with the stated purpose of test automation. - [INDIRECT_PROMPT_INJECTION]: The skill exhibits a potential surface for indirect prompt injection attacks.
- Ingestion points: The skill is triggered by user requests and interactions regarding test data and patterns.
- Boundary markers: The instructions do not define clear boundaries or delimiters to differentiate between system instructions and potentially malicious content within test files or data being processed.
- Capability inventory: The skill has powerful capabilities including file modification (Write, Edit) and full shell execution (Bash).
- Sanitization: There is no mention of sanitizing or validating external input before it is used in command execution or file writes.
Audit Metadata