magic-hour-explainer-video

Pass

Audited by Gen Agent Trust Hub on Nov 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided source documents and briefs to generate scripts and video content, which is an ingestion point for untrusted data.
  • Ingestion points: The agent is instructed to "Read the actual source material" and "Convert the brief into a clear story" (SKILL.md).
  • Boundary markers: The instructions do not specify the use of delimiters or explicit boundary markers to isolate the user-provided content from the agent's instructions.
  • Capability inventory: The skill utilizes specialized tools for media generation, including ai_voice_generator_create_audio, ai_voice_cloner_create_audio, ai_image_generator, and create_video_from_image. It does not possess capabilities for arbitrary command execution or system file modification.
  • Sanitization: No specific sanitization or filtering of the source material is mentioned prior to its use in script generation or tool invocation.
  • [EXTERNAL_DOWNLOADS]: The skill references external URLs for connecting to the Magic Hour Model Context Protocol (MCP) server and accessing connection documentation.
  • Evidence: The skill points to https://mcp.magichour.ai/ and a quickstart guide on GitHub (SKILL.md).
  • Context: These references target the official infrastructure and repositories of the skill author (magichourhq) and are necessary for the skill's intended functionality.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 5, 2026, 10:58 PM