Ozor Ozor is an AI video creation platform that turns ideas, documents, prompts, and URLs into branded videos with voiceover, scene-by-scene structure, and a polished editor. Instead of stitching together separate tools for scripting, design, voice, and rendering, Ozor handles the whole pipeline end to end. You give it a brief or a source file, the agent plans the scenes, generates visuals and narration, and you get a ready-to-publish video with full editorial control. It's built for teams that need video at the pace of writing: marketing, internal training, product launches, course creation, and outbound campaigns. Outputs are exportable as MP4 in 720p, 1080p, or 4K, and shareable via Private Exports with viewer tracking. The Ozor MCP The Ozor MCP is the Model Context Protocol connector that brings Ozor directly into any AI assistant that speaks MCP, including Claude. It removes the context switch entirely. You stay in your assistant, describe what you want, and the assistant calls Ozor for you. What it exposes: generate_video — create a new video from a short text prompt, with aspect ratio and export options analyze_document and generate_from_plan — turn a PDF, PPTX, DOCX, or URL into a scene-by-scene plan, edit it, then render send_message — give natural language edit instructions to an existing video (change copy, swap colors, reorder scenes, add a logo) export_video and wait_for_export — kick off and track MP4 renders list_videos, get_video, get_plan — inspect what you've created list_voices — pick from the available TTS voices for narration get_embed_code — get ready-to-use embed snippets for finished videos The workflow becomes conversational: "Make a 20 second portrait video promoting our new feature" → the assistant generates, you preview in the Ozor editor, ask for tweaks in chat, and export when you're happy. No tab switching, no copying briefs between tools, no leaving the surface where you already think. One connector, full video creation, where your AI already lives.
How to connect
https://server.smithery.ai/ozor/ozorvideo/mcpcurl -X POST https://server.smithery.ai/ozor/ozorvideo/mcp \
-H 'Content-Type: application/json' \
-H 'Accept: application/json, text/event-stream' \
-d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{}}'