chat-with-anyone
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
subprocess.runinscripts/extract_ref_segment.pyto callffmpeg. This is a legitimate and necessary use case for extracting audio segments from video files and does not involve untrusted user input in the command structure. - [EXTERNAL_DOWNLOADS]: The skill downloads media from YouTube and Bilibili using the well-known
yt-dlptool. This is the core functionality of the skill and follows safe usage patterns. - [DATA_EXFILTRATION]: The skill communicates with
noiz.ai(the vendor's domain) to perform voice design and synthesis tasks. It correctly manages theNOIZ_API_KEYby reading from standard environment variables or a local config file (~/.noiz_api_key), adhering to security best practices for secret management.
Audit Metadata