to-questionnaire
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to create and write a Markdown file named
to-questionnaire-<slug>.mdin the current working directory. This is a standard file-system operation for document generation and does not involve executing arbitrary system commands or scripts. - [INDIRECT_PROMPT_INJECTION]: The skill functions by ingesting untrusted input from the user (role, expertise, specific facts/decisions) and interpolating it into a predefined Markdown template.
- Ingestion points: User input provided during the interactive 'interview' phase.
- Boundary markers: None explicitly defined in the instructions, but the output is strictly restricted to a Markdown document structure.
- Capability inventory: The skill only performs file-write operations for the generated Markdown.
- Sanitization: There is no explicit sanitization of user-provided content before it is written to the file, but the impact is limited to the text content of the generated document.
Audit Metadata