movie_edit MCP for AI video apps and creative agents

Seedance Video MCP

Add prompt-to-video, media-reference generation, async status tracking, and shareable video artifacts to your AI product without building the provider workflow yourself.

  • check_circle Create video jobs from prompts plus image, video, or audio references
  • check_circle Poll normalized job state until queued, running, succeeded, or failed
  • check_circle Return safe preview artifacts and public share links for completed videos
Seedance Video MCP icon

Video job

Generated creative pipeline

Prompt

Create a 9:16 product launch video using @Image1 as the subject reference.

Status

running

Output

video/mp4

Artifacts

Preview URL, artifact ID, MIME type, and share link when ready.

Tools

Seedance Video exposes four public tools. Together they give your agent a reliable async video workflow from creation through sharing.

movie

create_video_generation_job

Use when the user has approved a production prompt and optional references. It creates an async Seedance job.

Required

prompt

Optional

references, settings

Returns

structuredContent.job

sync

get_video_generation_job

Use this polling tool after creation. It returns the current safe job payload until the job reaches a terminal state.

Required

job_id

Status

queued, running, succeeded, failed

Returns

job.artifacts, job.error

history

list_video_generation_jobs

Use when the user asks for recent generations, prior drafts, or active jobs tied to the authenticated Infoseek end user.

Required

None

Optional

limit from 1 to 50

Returns

structuredContent.jobs

ios_share

share_video_generation_artifact

Use after a job succeeds to create a public share link for a completed video artifact owned by the authenticated user.

Required

job_id

Optional

artifact_id

Returns

structuredContent.share

Response JSON

The MCP preserves the fields an AI app needs for continuity: stable job IDs, status, settings, references, artifacts, and safe error text.

job_id

Stable key for polling, follow-up turns, and artifact sharing.

references

Carries image, video, or audio references with assigned handles such as @Image1.

artifacts

Contains preview-safe video outputs your UI or agent can render.

settings

Keeps duration, ratio, resolution, audio, and watermark choices explicit.

{
  "structuredContent": {
    "job": {
      "job_id": "job_8c1f4b7d2a9e0f31",
      "status": "succeeded",
      "operation": "video_generation",
      "prompt": "Create a 9:16 launch video using @Image1...",
      "references": [
        {
          "kind": "image",
          "source_url": "https://infoseek.ai/tmp/ref/product.png",
          "display_name": "product.png",
          "mime_type": "image/png",
          "handle": "@Image1"
        }
      ],
      "settings": {
        "duration": 5,
        "ratio": "9:16",
        "resolution": "720p",
        "generate_audio": false,
        "watermark": false
      },
      "artifacts": [
        {
          "artifact_id": "art_1",
          "kind": "video",
          "preview_url": "https://infoseek.ai/tmp/output/video.mp4",
          "mime_type": "video/mp4"
        }
      ]
    }
  },
  "content": [
    { "type": "text", "text": "Video generation finished." }
  ]
}

Quickstart + Code

A production integration should create once after user approval, then poll the returned job ID until a terminal status is reached.

Async workflow

  1. 1. Build a reviewed prompt and align any reference handles with the references array.
  2. 2. Call create_video_generation_job once and persist the returned job_id.
  3. 3. Poll get_video_generation_job until the job succeeds or fails.
  4. 4. Optionally call share_video_generation_artifact for a public share link.
const created = await mcp.callTool("create_video_generation_job", {
  prompt: "Create a 9:16 social video using @Image1 as the product reference.",
  references: [
    {
      kind: "image",
      source_url: "https://infoseek.ai/tmp/ref/product.png",
      display_name: "product.png",
      mime_type: "image/png",
      handle: "@Image1"
    }
  ],
  settings: {
    ratio: "9:16",
    duration: 5,
    resolution: "720p",
    generate_audio: false,
    watermark: false
  }
});

const jobId = created.structuredContent.job.job_id;
const latest = await mcp.callTool("get_video_generation_job", { job_id: jobId });

Access

Every advertised Seedance Video tool requires Infoseek end-user OAuth. The Infoseek MCP route is the supported integration surface.

shield_lock

OAuth-Protected Tools

Use Infoseek auth when your AI app needs user-owned video jobs, private artifact handling, and continuity across follow-up turns.

movie_filter

Safe Artifact Surface

The public response omits private provider task IDs, private storage keys, and billing internals while preserving useful preview fields.

Why These Outputs Matter

Agent Continuity

Stable job IDs and sanitized job payloads let agents continue from the last real video state instead of guessing from prose.

Media-Aware Prompts

Reference handles keep image, video, and audio inputs aligned with prompt instructions for better creative control.

UI-Ready Artifacts

Preview URLs and artifact IDs give your app a direct path from generation status to playback and sharing.

Use Cases

campaign

Creative Automation

Let marketing agents generate short social variations from product images, motion references, and approved campaign prompts.

smart_display

AI Video Editors

Build chat-based editing workflows that use video references, image replacements, duration matching, and status-aware polling.

auto_awesome

Generated Media Products

Add video generation to creator tools, ad builders, and internal brand studios with a clean MCP contract.