teach
Pass
Audited by Gen Agent Trust Hub on Jul 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions direct the agent to open generated HTML lesson files using system CLI commands (such as 'open' or 'start') to present content to the user.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface due to its reliance on external data:
- Ingestion points: The agent reads and processes data from local files (MISSION.md, RESOURCES.md) and is explicitly instructed to fetch knowledge from external high-trust URLs rather than relying on internal knowledge.
- Boundary markers: There are no instructions provided to use delimiters or ignore embedded instructions when reading these files or external resources.
- Capability inventory: The skill performs file system writes (lessons, records) and shell command execution (opening files).
- Sanitization: The skill lacks instructions for validating or sanitizing external content before it is incorporated into the learning workspace or agent context.
Audit Metadata