deep-learn
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security risks were identified. The skill focuses on managing a learning session through markdown documentation and user interaction.
- [COMMAND_EXECUTION]: The skill instructs the agent to read relevant source code files and write/update a local markdown checklist (
deep-learn-<topic-slug>.md). These are standard file operations necessary for the skill's stated purpose and do not represent a privilege escalation or unsafe execution risk. - [PROMPT_INJECTION]: The skill includes instructions to guide agent behavior (e.g., 'never declare mastery the learner hasn't shown'), but these are aligned with the skill's primary function as a tutor and do not attempt to bypass core system safety filters or extract system prompts.
- [DATA_EXFILTRATION]: No network operations or external data transmissions are present. The skill interacts solely with the local filesystem (writing a checklist) and the user (via
AskUserQuestion). - [EXTERNAL_DOWNLOADS]: The skill does not perform any external downloads or package installations.
Audit Metadata