teach
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external content such as source code, git diffs, and conversation history to explain concepts to the user. This creates a surface where malicious instructions embedded in the analyzed code or documentation could attempt to influence the agent's behavior.
- Ingestion points: The skill reads codebase content and diffs to orient itself and provide explanations (SKILL.md).
- Boundary markers: The instructions do not define specific delimiters or instructions to ignore commands potentially hidden within the analyzed source material.
- Capability inventory: The skill coordinates other tools (how, why, unslop), generates diagrams, and interacts with development tools like the debugger and code viewer.
- Sanitization: The skill does not implement specific filtering or sanitization of the input data before processing.
- [COMMAND_EXECUTION]: The instructions direct the agent to interact with development environment tools, including opening code files, viewing diffs, and utilizing the debugger to assist in explanations. These capabilities are consistent with the skill's primary purpose as a developer assistant but involve active interaction with the user's local environment.
Audit Metadata