to-questionnaire

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user input to populate a markdown questionnaire.
  • Ingestion points: User input identifying recipient roles, expertise, and specific fact-finding needs (SKILL.md).
  • Boundary markers: Absent; the process does not specify the use of delimiters to segregate user input from instructions.
  • Capability inventory: File-writing capability to create markdown files in the .agents/projects/ path.
  • Sanitization: Absent; no explicit sanitization of user-provided strings is requested before file generation.
  • [SAFE]: The skill enforces project-root path resolution for all file operations, effectively preventing path traversal or accidental modification of sensitive global agent files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 01:26 PM
Security Audit — agent-trust-hub — to-questionnaire