kling-video
Pass
Audited by Gen Agent Trust Hub on Jul 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security vulnerabilities were detected. The skill follows security best practices for API integration and data handling.- [EXTERNAL_DOWNLOADS]: The skill downloads media assets from approved Kling AI delivery hosts. The implementation includes defensive measures such as DNS resolution pinning and explicit host allowlisting to prevent interaction with malicious domains.- [COMMAND_EXECUTION]: The provided Python script uses hardcoded paths to call standard media utilities (ffprobe, ffmpeg) for integrity validation. These calls are performed without shell execution, mitigating risks of command injection.- [CREDENTIALS_UNSAFE]: The skill correctly handles sensitive credentials by recommending the use of environment variables and explicitly cautioning against exposing API keys in client-side code.
Audit Metadata