teach
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The agent is instructed to open lesson files using CLI commands, which is a functional convenience for the tutor role.
- [DYNAMIC_EXECUTION]: The skill involves generating interactive HTML documents with potential JavaScript components for quizzes and simulators.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-controlled files as state, creating an indirect injection surface.
- Ingestion points: MISSION.md, RESOURCES.md, learning-records/, and NOTES.md.
- Boundary markers: The instructions do not define delimiters to separate user data from agent instructions.
- Capability inventory: The agent can perform file writes and CLI command execution.
- Sanitization: There are no explicit requirements for sanitizing content sourced from external resources or user files.
Audit Metadata