teach
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes information from various workspace files that may contain untrusted data from external resources or user input. This data is used to calculate the teaching strategy and generate content. \n
- Ingestion points: Data is read from MISSION.md, RESOURCES.md, learning-records/*.md, and NOTES.md. \n
- Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the ingested files. \n
- Capability inventory: The agent can write files (HTML lessons and markdown records) and execute CLI commands to open files. \n
- Sanitization: There is no mention of sanitizing or validating content retrieved from external resources before it is used in the workspace. \n- [COMMAND_EXECUTION]: The skill instructions suggest the use of system CLI commands to facilitate the opening of generated lesson files. \n
- Evidence: SKILL.md contains the instruction: 'If possible, open the lesson file for the user by running a CLI command.'
Audit Metadata