Skip to content
Scalekit Docs
Talk to an Engineer Dashboard

Connect AI agents to the Runway MCP server

Vendor MCP
Open markdown

The Runway MCP connector routes your AI agent's tool calls to Runway's own MCP server through Scalekit. Each user signs in to Runway once, and Scalekit stores and refreshes their tokens, so your agent never handles credentials. It comes with 48 tools.

Tools
48
What they doRead · write · destructive
15 · 32 · 115 read32 write1 destructive
Users sign in with
OAuth app
Your own Runway MCP server app

Setup

  1. Install the SDK

    Terminal window
    npm install @scalekit-sdk/node dotenv
  2. Set your credentials

    Add your Scalekit credentials to your .env file. Find values in app.scalekit.com > Developers > API Credentials.

    .env
    SCALEKIT_ENVIRONMENT_URL=<your-environment-url>
    SCALEKIT_CLIENT_ID=<your-client-id>
    SCALEKIT_CLIENT_SECRET=<your-client-secret>
  3. Create the Runway MCP connection

    In AgentKit > Connections, create a Runway MCP connection and copy its redirect URI. The name you give it is the connection_name your code passes. See Configure connections.

  4. Register an OAuth app

    Runway MCP server connections use your own OAuth app. Register one with Runway MCP server and add the redirect URI you copied.

    Then enter the app's Client ID and Client Secret on the Runway MCP connection.

  5. Authorize a user and make your first call

    quickstart.mts
    import { ScalekitClient } from '@scalekit-sdk/node'
    import 'dotenv/config'
    import { createInterface } from 'node:readline/promises'
    const scalekit = new ScalekitClient(
    process.env.SCALEKIT_ENVIRONMENT_URL,
    process.env.SCALEKIT_CLIENT_ID,
    process.env.SCALEKIT_CLIENT_SECRET,
    )
    const actions = scalekit.actions
    const connector = 'runwaymcp'
    const identifier = 'user_123'
    // Generate an authorization link for the user
    const { link } = await actions.getAuthorizationLink({ connectionName: connector, identifier })
    console.log('Authorize Runway MCP:', link)
    const rl = createInterface({ input: process.stdin, output: process.stdout })
    await rl.question('Press Enter after authorizing...')
    rl.close()
    // Make your first call
    const result = await actions.executeTool({
    connector,
    identifier,
    toolName: 'runwaymcp_whoami',
    toolInput: {},
    })
    console.log(result)
    Terminal window
    npx tsx quickstart.mts

    Each user signs in once. See Authorize a user for the full flow and statuses.

Tools

Pass the exact name to execute_tool
Try in PlaygroundRequest a tool
  • runwaymcp_get_agent_runCheck a Runway Agent run, waiting up to wait_seconds for it to need the user, finish, or report new render progress.Read-only

    Check an Agent Run

    Check a Runway Agent run, waiting up to wait_seconds for it to need the user, finish, or report new render progress. Keep checking this tool while the run is working, even if the inline card is updating: the card cannot ask questions in chat. Let the card display progress without repeating its status narration. When inputRequired is true: show the user the message from Runway and give them each available preview runwayUrl as a link to review it in Runway. If linksUnavailable is true, retry get_agent_run once, then explain if links remain unavailable. The url field is signed media for inline display only. Then ask every question in inputRequests with your question UI, offering its options (the default is Runway's recommendation; the user may instead type their own answer, which goes in the field ending in __own). Before an approval question, briefly summarize the proposed creative and any stated credit cost; do not invent missing details. Keep tool names, run ids, and process instructions out of your user-facing prose. Send the user's answers with answer_agent_run. Never answer or approve for the user, and do not skip a question. When status is SUCCEEDED, call show_agent_run only if the finished ad is not already displayed in a card in this chat turn. This brings a result shown only in an earlier turn beside your final response without duplicating a current card. Give the user each available deliverable runwayUrl as a link, with its label and post text. While the run is working, call this again. Without an inline card, also give one short progress update.

    Inputs

    run_idstringrequired
    The run id create_ad returned.
    wait_secondsnumber
    How long to wait for news, up to 45 seconds. Default 40.
  • runwaymcp_get_brand_kitGets one visible Brand Kit from the connected workspace, including its curated image, video, and audio assets grouped by category.Read-only

    Get Brand Kit

    Gets one visible Brand Kit from the connected workspace, including its curated image, video, and audio assets grouped by category. Call this when a kit has been selected, when there is one clearly applicable kit, or when the user asks what is inside a kit. Image and video asset URLs can be passed unchanged into the existing generation tools as references. Reading Brand Kits is free and non-destructive.

    Inputs

    brand_kit_idstringrequired
    Brand Kit UUID from list_brand_kits.
    rationalestring
    Why are you calling this tool? What is the user actually trying to accomplish? Be specific — e.g., "user uploaded a vacation photo and asked to animate it as a 5-second clip for Instagram".
  • runwaymcp_get_skillLoads a Runway skill: step-by-step instructions for a creative task that combines several Runway tools.Read-only

    Get Runway Skill

    Loads a Runway skill: step-by-step instructions for a creative task that combines several Runway tools. When the user's request matches a skill below, call this before any generation tool and follow the returned instructions instead of generating directly. Skip it when the skill is already loaded in this conversation. Reading a skill is free. Skills: - `mood-board`: Create a mood board that sets the visual direction for a project — use when the user asks for a mood board, or wants to see a collection of reference images presented together in a single composed output rather than as separate standalone images. - `resize-image`: Change an existing image's aspect ratio — resize, reframe, or extend it to a new ratio (like 21:9, 1:1, or 9:16) while keeping the original content intact and filling the new areas with contextually relevant extensions. Not for generating a new image from a description or for cropping content out. - `social-thumbnail`: Create a thumbnail or cover image for social media — a single still designed to grab attention with a readable headline and at most one line of subtext. Not for motion video or multi-shot pieces (use generate_video or generate_multishot_video), nor for plain illustrative stills with no headline or layout intent (use generate_image directly). - `workflow-builder`: Create, edit, inspect, or run a node-based Runway workflow — a reusable pipeline of generation steps ('build a workflow that makes a 1-minute product ad video', 'a workflow that restyles my static ad with different characters', 'run my workflow'). Use when the deliverable is the reusable workflow itself, or a run of one, not a single hand-orchestrated output.

    Inputs

    namestringrequired
    Skill name from the list in this tool description.one of mood-boardresize-imagesocial-thumbnailworkflow-builder
    rationalestring
    Why are you calling this tool? What is the user actually trying to accomplish? Be specific — e.g., "user uploaded a vacation photo and asked to animate it as a 5-second clip for Instagram".
  • runwaymcp_get_taskGets details for a Runway task by ID, or generation details for an authorized asset by asset_id.Read-only

    Check Task Status

    Gets details for a Runway task by ID, or generation details for an authorized asset by asset_id. Generation tools (`generate_image`, `generate_video`, `generate_multishot_video`, `run_workflow`) return immediately with a task ID. If an inline viewer rendered, do not poll; the viewer updates itself. If no viewer rendered, poll this tool until SUCCEEDED (wait 60-120s between calls for video/workflow). For `dynamic_workflow` tasks, the text includes overall %, per-step name/status, and output URLs when nodes finish. Parent `dynamic_workflow` SUCCEEDED can have empty artifacts — use the step Outputs, not artifacts[0]. Set `include_workflow_execution: true` only when you need node UUIDs. When status is SUCCEEDED the response includes the asset URL. USER-FACING REPLIES: describe the asset content; the `taskType` in the response encodes the model name and is for debugging only — keep it out of your reply unless the user explicitly asks which model was used.

    Inputs

    asset_idstring
    Asset UUID. Returns generation details for this selected asset; use instead of id.
    idstring
    Task ID. Provide exactly one of id or asset_id.
    include_workflow_executionboolean
    Workflow runs only (`run_workflow`). If true, also include node UUIDs. Ignored for other tasks. Default is named steps + output URLs.
    rationalestring
    Why are you calling this tool? What is the user actually trying to accomplish? Be specific — e.g., "user uploaded a vacation photo and asked to animate it as a 5-second clip for Instagram".
  • runwaymcp_get_workflowGets a Runway workflow by id (name, description, updated time).Read-only

    Get Workflow

    Gets a Runway workflow by id (name, description, updated time). Set `include_graph: true` to also load the saved graph. Optional `version` loads a specific version instead of latest. The graph JSON is for your next tool call — summarize the pipeline for the user; do not dump node IDs or ASCII diagrams unless they ask how it is wired. Do not invent graph JSON. `list_workflows` → `get_workflow` include_graph: true and copy nodes/edges. Read `runway://docs/workflows/authoring` before composing or changing a graph, `runway://docs/workflows/examples` for the API shape, and `runway://docs/workflows/models` before image/video chains.

    Inputs

    workflow_idstringrequired
    Workflow UUID from list_workflows, create_workflow, or save_workflow_version.
    include_graphboolean
    If true, include the saved graph (latest, or `version` if set).
    rationalestring
    Why are you calling this tool? What is the user actually trying to accomplish? Be specific — e.g., "user uploaded a vacation photo and asked to animate it as a 5-second clip for Instagram".
    versionnumber
    Specific version number to load when include_graph is true (default: latest).
  • runwaymcp_list_brand_kitsLists the connected workspace's visible Brand Kits with their category names and descriptions, without loading assets.Read-only

    List Brand Kits

    Lists the connected workspace's visible Brand Kits with their category names and descriptions, without loading assets. Use this for cheap discovery when the user asks for branded work and no Brand Kit is already in context. Call `get_brand_kit` only after a kit is selected or when there is a single clearly applicable kit. An empty list means this workspace has no visible Brand Kits.

    Inputs

    rationalestring
    Why are you calling this tool? What is the user actually trying to accomplish? Be specific — e.g., "user uploaded a vacation photo and asked to animate it as a 5-second clip for Instagram".
  • runwaymcp_list_recentLists recent uploaded and generated assets for the authenticated workspace.Read-only

    Runway

    Lists recent uploaded and generated assets for the authenticated workspace. Returns asset IDs, media types, task IDs when available, and reusable asset URLs. Use this to help the user find recent images, videos, and audio.

    Inputs

    limitnumber
    Max results (10-50, default 12).
    rationalestring
    Why are you calling this tool? What is the user actually trying to accomplish? Be specific — e.g., "user uploaded a vacation photo and asked to animate it as a 5-second clip for Instagram".
  • runwaymcp_list_workflowsLists Runway editor workflows in the authenticated workspace (name, id, updated time).Read-only

    List Workflows

    Lists Runway editor workflows in the authenticated workspace (name, id, updated time). Does not include the graph — use `get_workflow` with `include_graph: true` to inspect. If a similar workflow already exists, reuse it (`get_workflow` / `save_workflow_version` / `run_workflow`) instead of creating a duplicate. Do not invent graph JSON. `list_workflows` → `get_workflow` include_graph: true and copy nodes/edges. Read `runway://docs/workflows/authoring` before composing or changing a graph, `runway://docs/workflows/examples` for the API shape, and `runway://docs/workflows/models` before image/video chains.

    Inputs

    cursorstring
    Pagination cursor from a previous list_workflows response.
    limitnumber
    Max results (10-100, default 12).
    rationalestring
    Why are you calling this tool? What is the user actually trying to accomplish? Be specific — e.g., "user uploaded a vacation photo and asked to animate it as a 5-second clip for Instagram".