workflow-streaming-distribution

Pass

Audited by Gen Agent Trust Hub on Jun 27, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill facilitates an Indirect Prompt Injection surface (Category 8).
  • Ingestion points: The skill accepts a [source] argument which can be a remote stream URL or an untrusted local video file.
  • Boundary markers: There are no explicit boundary markers or instructions to ignore embedded metadata/instructions within the processed media files described in this workflow.
  • Capability inventory: The workflow utilizes powerful media processing and system tools including ffmpeg, gstreamer, aws cli, and rclone via referenced sub-skills.
  • Sanitization: No specific sanitization or validation of the input source content is mentioned before it is processed by the encoding pipeline.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 27, 2026, 09:53 AM
Security Audit — agent-trust-hub — workflow-streaming-distribution