ai-for-learning
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [Identity and Purpose] (SAFE): The skill defines a persona for an AI Learning Architect focused on educational technology. All instructions are focused on providing expertise in AI tutoring, personalization, and content generation.- [Reference System] (SAFE): The skill uses local markdown files for grounding knowledge. There are no external URLs or untrusted remote data sources involved.- [No Executable Code] (SAFE): Analysis of all files (SKILL.md, patterns.md, sharp_edges.md, validations.md) confirms the absence of scripts, binaries, or shell commands. The content is purely text-based documentation and prompt templates.- [Indirect Prompt Injection] (LOW): While the skill describes processing student work or course content (untrusted data), it does not provide autonomous tool capabilities that could be exploited. The risk is limited to the AI's response content within its primary advisory role.
Audit Metadata