teach
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from external resources listed in
RESOURCES.mdand use them to construct lessons. - Ingestion points:
RESOURCES.md, external websites, and primary sources (videos/articles) referenced during the teaching process. - Boundary markers: The instructions lack explicit delimitation or instructions for the agent to ignore embedded commands within the external materials it processes.
- Capability inventory: The skill has the ability to write multiple file types (
.md,.html, and assets) to the workspace and execute shell commands via the CLI. - Sanitization: There is no requirement or logic provided to sanitize or validate external content before it is interpolated into lessons or processed by the agent.
- [COMMAND_EXECUTION]: The instructions explicitly direct the agent to "open the lesson file for the user by running a CLI command," which involves executing shell commands based on the filenames it has generated.
- [DYNAMIC_EXECUTION]: The skill generates interactive components such as "quiz widgets," "simulators," and "diagram helpers" which involve writing and subsequently executing HTML, CSS, and potentially Javascript in the user's local environment.
Audit Metadata