learn-codebase
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze unfamiliar codebases, which serve as untrusted external data sources. Malicious instructions hidden within the source code or comments of the codebase being studied could attempt to influence the agent's behavior.
- Ingestion points: File reads from the target repository being learned (SKILL.md).
- Boundary markers: The instructions do not specify any delimiters or warnings to ignore instructions found within the code being analyzed.
- Capability inventory: The skill requires writing files (MISSION.md, NOTES.md, lessons/), modifying code (adding console logs), and executing code to verify behavior.
- Sanitization: There are no instructions for sanitizing or validating the content of the repository before interaction or execution.
- [COMMAND_EXECUTION]: The skill encourages a "prove-it" phase where the agent is expected to edit and run the target codebase. While this is the intended purpose of the skill, it involves executing code from an unfamiliar source which could contain malicious logic.
Audit Metadata