game-video-ad-pipeline
Pass
Audited by Gen Agent Trust Hub on May 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The script
scripts/generate-video.pyinvokesffmpegandffprobeusing list-based arguments with thesubprocessmodule. This approach prevents shell injection vulnerabilities. Critical inputs like the Steam game ID are validated to be strictly numeric before use. - [EXTERNAL_DOWNLOADS]: The skill downloads game metadata and screenshots from
steampowered.com, a well-known and trusted service. The network operations use the standardurllib.requestlibrary with defined timeouts and appropriate headers. - [SAFE]: The skill does not exhibit any malicious behaviors such as data exfiltration, persistence, or obfuscation. Text from external sources (Steam storefronts) is safely rendered into static images via the Pillow library before being overlaid on the video, providing strong isolation against potential injection attacks in the video processing pipeline.
Audit Metadata