magic-hour-text-to-video

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill facilitates downloading video files from URLs returned by the Magic Hour API. These resources are hosted on infrastructure controlled by the skill's author (magichour.ai).
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted user-provided text to generate video content. 1. Ingestion points: User-provided ideas or prompt descriptions processed in SKILL.md. 2. Boundary markers: The skill does not define specific delimiters for user input, relying on the agent to interpret the idea correctly. 3. Capability inventory: The skill uses tools to create video projects, poll for status, and download final assets. 4. Sanitization: No explicit programmatic sanitization is defined; the skill instructs the agent to Translate the idea which involves manual re-interpretation.
  • [COMMAND_EXECUTION]: The skill utilizes Model Context Protocol (MCP) tools for video generation. These tools represent the core functionality of the skill and are operated within the agent's authorized toolset.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 08:30 AM
Security Audit — agent-trust-hub — magic-hour-text-to-video