edu-math-tutorial

Pass

Audited by Gen Agent Trust Hub on Sep 1, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input (math problems) and uses it to construct instructions for downstream AI agents and rendering tools.
  • Ingestion points: The skill ingests a user-provided math problem to extract the '题面' (problem text) and '解法链' (solution chain) as described in Section 1.
  • Boundary markers: There are no explicit instructions or delimiters to isolate user-provided content from the system instructions when passed to the ai_editing_director.
  • Capability inventory: The skill has the capability to delegate tasks to the ai_editing_director and design_motion_overlays tools (Section 5) to create motion clips, audio scripts, and overlays.
  • Sanitization: There is no mention of filtering, escaping, or validating the input math problem text, which could allow a malicious user to embed instructions within a problem description to influence the video generation process.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 1, 2026, 10:35 AM
Security Audit — agent-trust-hub — edu-math-tutorial