to-questionnaire
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists of instructions to interview the user and write a file using a static template. No malicious patterns such as credential harvesting, remote code execution, or data exfiltration were found.\n- [INDIRECT_PROMPT_INJECTION]: The skill provides an attack surface for indirect prompt injection as it ingests untrusted user data to generate files.\n
- Ingestion points: User input collected in Steps 1 and 2 regarding the recipient and the specific facts needed (SKILL.md).\n
- Boundary markers: None. The skill lacks specific instructions to delimit user input or prevent the agent from obeying embedded instructions in the user's responses.\n
- Capability inventory: Local file write capability (
to-questionnaire-<slug>.md).\n - Sanitization: Absent. There is no instruction to escape or validate user-provided strings before they are used in the generated file or its name.
Audit Metadata