magic-hour-video-editing

Pass

Audited by Gen Agent Trust Hub on Sep 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The instructions in SKILL.md direct the agent to utilize available local command-line tools such as FFmpeg for non-generative tasks like trimming, resizing, and reframing footage.
  • [EXTERNAL_DOWNLOADS]: The skill in SKILL.md references external documentation and setup guides hosted on the vendor's official domain (magichour.ai) and GitHub organization (magichourhq).
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied video files which represents a potential attack surface for indirect prompt injection via media metadata or embedded data.
  • Ingestion points: External video files and metadata (audio streams, duration, dimensions) accessed as described in SKILL.md.
  • Boundary markers: Absent; the instructions do not specify delimiters to separate untrusted media metadata from agent instructions.
  • Capability inventory: Network access to Magic Hour API endpoints and shell execution for video processing tools (FFmpeg) across scripts.
  • Sanitization: The skill does not describe specific validation or sanitization steps for data extracted from video files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 12, 2026, 05:09 PM
Security Audit — agent-trust-hub — magic-hour-video-editing