Teach
Pass
Audited by Gen Agent Trust Hub on Aug 15, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
curlto send POST requests tohttp://localhost:31337/notify. This is a standard mechanism for local inter-process communication, specifically for triggering a Text-to-Speech (TTS) engine. The commands are fire-and-forget and do not ingest untrusted remote data for execution. - [DATA_EXFILTRATION]: While the skill performs network requests, they are strictly constrained to
localhost. There is no evidence of data being sent to external domains or sensitive files being accessed. - [PROMPT_INJECTION]: The instructions include clear boundaries for how to handle user input (e.g., '/teach-me' arguments) and maintain a structured Socratic dialogue. It does not attempt to bypass safety filters or override system-level instructions.
- [REMOTE_CODE_EXECUTION]: No external scripts are downloaded or executed. The dependencies are limited to the local system's
curlutility and a presumed local voice server.
Audit Metadata