note

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local Python scripts, specifically look_tongji.py and timeline_tools.py, located within the repository's internal structure (<SKILL_DIR>/../../scripts/). These scripts handle the transcription and processing of lecture data.
  • [EXTERNAL_DOWNLOADS]: As part of its core functionality, the skill uses a command-line interface to download lecture videos, transcripts, and slides. These operations are initiated via local scripts and target academic/course content platforms.
  • [SAFE]: The instructions include strict formatting rules and "FORBIDDEN" phrases (e.g., meta-commentary about the note-taking process) which serve as a safeguard against the agent including extraneous information or following instructions embedded within the processed transcripts (Indirect Prompt Injection surface).
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 11:56 AM
Security Audit — agent-trust-hub — note