teach
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's fundamental operation involves reading and synthesizing information from user-controlled files such as MISSION.md, RESOURCES.md, and learning records. This creates a potential surface for indirect prompt injection where instructions hidden within these files could attempt to manipulate the agent's behavior during lesson creation or feedback sessions.\n
- Ingestion points: The agent ingests data from MISSION.md, learning-records, RESOURCES.md, NOTES.md, and various HTML files in the reference directory.\n
- Boundary markers: The instructions lack specific delimiters or warnings to the agent to distinguish between its own logic and data retrieved from the workspace files.\n
- Capability inventory: The agent has the ability to write files (Markdown and HTML) to the workspace and is encouraged to execute CLI commands to open these files for the user's convenience.\n
- Sanitization: There are no outlined procedures for validating or sanitizing the content extracted from these files prior to its use in generating further instructions or lessons.
Audit Metadata