The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data, including video transcripts, footage, and outlines, which could contain adversarial instructions designed to influence the agent's behavior during the analysis or generation phase.
Ingestion points: Video transcripts, outlines, footage, and source images provided by the user.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the source material.
Capability inventory: The skill utilizes image generation tools (ai_image_editor_create_image, ai_image_generator_create_image) and network upload capabilities (video_assets_generate_presigned_url).
Sanitization: No explicit sanitization or validation of the input text/metadata is described.
[EXTERNAL_DOWNLOADS]: The skill references the vendor's official MCP endpoint and documentation repositories for configuration and API usage.
Evidence: References to https://mcp.magichour.ai/ and the magichourhq GitHub organization.

magic-hour-thumbnails