Integration

Ozor integration

Ozor (Mintii Labs) turns prompts and documents into animated, narrated videos. Generate a video from a prompt and edit it by instruction (its own AI agent builds the scenes), turn a PDF, PowerPoint, Word file or web page into a scene plan, edit the plan and render it, export MP4s at 720p, 1080p or 4K (plan-dependent), and list videos and narration voices. A prompt video or an edit costs 1 credit when it succeeds, a video from a plan max(1, floor(narrated scenes / 2)) credits, and exports are free; a public export gets a permanent link anyone can open. Left out: the API key management routes (they need the browser session), the agent tool-trace diagnostic (no route returns its conversation id) and the MCP server. Authentication is an Ozor API key (sk_live_...) created in Settings, Developer, API Keys, sent in the X-API-Key header.

Connect Ozor How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Ozor.Built from the 11 Ozor tools below, on the connection you add once.

What you can do with Ozor

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Each badge shows the default agent permission of an endpoint (allowed: an agent calls it without asking; ask first: you are asked before an agent calls it, when the agent or chat asks permission for sensitive actions; never: no agent may call it), which every user can change endpoint by endpoint in their credential settings.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Generate, 2 operations

generate_videoGenerate videoPOSTAgents: ask firstStarts Ozor's agent building a new animated video from a prompt; answers videoId and jobId at once. Poll get_video_job until completed or failed. Costs 1 credit when the job succeeds, none if it fails.
generate_video_from_planGenerate video from planPOSTAgents: ask firstRenders a draft or failed plan into a video project, streaming progress; the last chunk (step done) carries projectId, the videoId for export_video. Costs creditsCost of the plan, at least 1; a short balance fails at once.

Plan, 2 operations

get_document_planGet document planGETAgents: ask firstReturns a plan: status, language, style, scenes (order, title, visual prompt, narration, images with fresh short-lived URLs, key points), creditsCost and, after generation, projectId. Read it right before editing.
update_document_planUpdate document planPATCHAgents: ask firstEdits a plan while it is draft or failed: replaces its scenes (send the complete list from get_document_plan with your changes; a scene left out is dropped) and/or sets the voice.

Video, 2 operations

get_videoGet videoGETAgents: ask firstReturns a video with its latest export: status, editorUrl, exportStatus (queued, processing, complete, failed), downloadUrl (signed, about 24 hours), shareUrl and shareCode when public, thumbnail and exportError.
list_videosList videosGETAgents: ask firstLists the videos the account created through the API (not those made in the web app), newest first: videoId, title, status, editorUrl and createdAt. Pass a videoId to get_video for its export and links.

Analyze, 1 operation

analyze_documentAnalyze documentPOSTAgents: ask firstTurns a PDF, PPTX or DOCX file (at most 50 MB) or a web page into a plan: scenes with narration script, visual description and extracted images, plus creditsCost, the credits generate_video_from_plan will take. Send file or url.

Export, 1 operation

export_videoExport videoPOSTAgents: ask firstRenders an MP4 of a video's current state and returns the video with downloadUrl (signed, about 24 hours) and, when public, shareUrl. No credit cost. Waits for the job and returns the finished result; gives up after 10 minutes.

Job, 1 operation

get_video_jobGet video jobGETAgents: ask firstReturns the status of an agent job from generate_video or send_video_message: queued, pending, processing, completed (response holds the agent's reply) or failed (error says why; no credit is charged). Poll every few seconds.

Message, 1 operation

send_video_messageSend video messagePOSTAgents: ask firstSends an edit instruction to Ozor's agent for an existing video (change copy, colors, scene order...); answers a jobId at once. Poll get_video_job until completed. Costs 1 credit when the job succeeds, none if it fails.

Voice, 1 operation

list_voicesList voicesGETAgents: ask firstLists the narration voices: id (aria, nova, felix, marcus, thor, luna, owen, sky), label, gender and character. Pass an id as voiceId to update_document_plan.

Use Ozor from Claude Code, Cursor or Codex

Connect Ozor in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Ozor operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Ozor

Create your workspace
Sign up free. Nothing to install.
Connect Ozor once
Add your Ozor connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.