MCQ Generator

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: No malicious patterns, prompt injections, or obfuscated instructions were found. The skill is dedicated to educational content generation and follows professional standards.\n- [NO_CODE]: The skill is composed of Markdown instructions and reference files. It does not contain or ship any executable code; the Python snippets in the reference documentation are static examples illustrating data formats and export logic.\n- [SAFE]: Evaluation of the Indirect Prompt Injection surface:\n
  • Ingestion points: Reference documents provided by the user (Phase 1, SKILL.md).\n
  • Boundary markers: None explicitly defined in the prompt text, relying on agent-level safety and the structured nature of questionnaire tasks.\n
  • Capability inventory: Web search (Phase 2) and generation of structured files (XML, GIFT, XLSX, DOCX) (Phase 5).\n
  • Sanitization: Strong mitigation via mandatory factual cross-referencing (Phase 2) and an explicit human-in-the-loop validation requirement (Phase 4) before any final output.\n- [SAFE]: The skill uses well-known educational and technical services (Moodle, Kahoot!, Wooclap, Word) and targets reputable information sources (Légifrance, academic sites) for its research protocol.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 07:33 PM
Security Audit — agent-trust-hub — MCQ Generator