tutor
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted markdown data from the user's directory to generate quiz content, which represents a potential surface for indirect prompt injection if notes contain hidden instructions.
- Ingestion points: Phase 3 reads markdown files from detected directories matching
**/StudyVault/to build questions. - Boundary markers: Absent; the skill does not use specific delimiters or instructions to ignore embedded commands within the ingested notes.
- Capability inventory: The skill has the capability to write and update local files (
concepts/{area}.mdand*dashboard*) and interact with the user viaAskUserQuestion. - Sanitization: No sanitization or validation of the markdown content is performed before it is interpolated into the quiz generation process.
Audit Metadata