legal-ai-course-editor
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external transcripts, creating a surface for indirect prompt injection. Evidence: 1. Ingestion points: transcripts read via file tools in SKILL.md Step 1. 2. Boundary markers: core principles 1 & 2 requiring fidelity to original text and no creation of new content. 3. Capability inventory: tool-based reading of files and writing of Markdown lecture notes to the filesystem. 4. Sanitization: lacks programmatic sanitization, relying on instructional constraints to filter content and mark uncertainties.
Audit Metadata