to-questionnaire
Pass
Audited by Gen Agent Trust Hub on Aug 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [SAFE]: The skill follows a standard interactive workflow to generate a Markdown file. No suspicious patterns such as network access, remote code execution, or obfuscation were detected.
- [COMMAND_EXECUTION]: The skill performs a file-write operation to save the generated questionnaire to the local filesystem (
to-questionnaire-<slug>.md). This behavior is transparent, documented, and restricted to the local directory, posing no threat. - [INDIRECT_PROMPT_INJECTION]: The skill provides a surface for data ingestion. 1. Ingestion points: User interview steps 1 and 2 in
SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: File-write (SKILL.mdstep 3). 4. Sanitization: Absent. As the data is provided directly by the user for document generation, this surface is considered safe. - [NO_CODE]: The skill contains no executable scripts or external code dependencies, relying solely on instructional prompts and a predefined document template.
Audit Metadata