web-material-sourcing
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a Python script (
scripts/download_material.py) to download external resources. This script is intended to be run by the agent to fetch verified materials and update a metadata manifest (materials.json). - [EXTERNAL_DOWNLOADS]: The skill uses
urllib.requestto fetch content from user-provided or search-derived URLs. It includes a safety check that limits downloads to 50MB per file and verifies HTTP status codes to prevent handling of error pages as valid data. - [SAFE]: The skill implements strong provenance tracking by requiring fields for license, source URL, and author for every downloaded item. It explicitly instructs the agent to respect robots.txt (via
WebFetch) and not to bypass paywalls or logins.
Audit Metadata