---
title: "Runway API MCP server for AI agents - actionlake"
description: "Give Claude 65 Runway API actions through one MCP server: recipes, voices, avatars. Your agent never holds the credential. 100,000 calls a month free."
canonical: https://actionlake.com/connectors/runway-api
updated: 2026-10-23
---

Connectors · AI

# Runway API MCP server for AI agents.

actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to Runway API and the rest of your systems. It gives your agent 65 Runway API actions (22 reads, 36 writes and 7 that wait for a confirmation) and holds the Runway API credential itself, so the agent never does. It paces Runway API's limits and records every call.

[Connect Runway API free](https://actionlake.com/console?start=sandbox) [See pricing](https://actionlake.com/pricing)

100,000 free calls a month. No card, no email, no password: press Start a free sandbox on the next screen and keep the key it shows you.

`claude mcp add --transport http actionlake https://connect.actionlake.com/mcp`

the first call

```
do {
  action: "runway.documents.list",
  intent: "check which Runway API account is connected"
}
# List documents · read · 1 upstream call
```

- 65 actions in one signed file
- 22 read look, never change
- 36 write create and update
- 7 wait for a yes destructive, confirmed first

Set up

## Connect Runway API in three steps.

1. 01
   ### Add actionlake to Claude.
   Run the command once in Claude Code; its consent page can create a free sandbox: 100,000 calls a month, no card. In the Claude app, add a custom connector with the URL below and paste your key on the consent page. Creating a new sandbox from the Claude app is not yet tested by us.
   `https://connect.actionlake.com/mcp`
2. 02
   ### Connect Runway API.
   Bearer token. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead. actionlake stores what you paste write-only, sealed under your tenant's key.
   - `api_key` Runway API secret key: in the Runway developer portal (https://dev.runwayml.com) open your organization > API Keys >…
   ```
   account { op: "connections.requirements", service: "runway" }
   # free: exactly what to paste, and what is found for you
   ```
3. 03
   ### Make the first call. Check before a write.
   Ask in plain words; Claude picks the action. Before anything changes, `check` validates the call and prices it, for free.
   ```
   do    { action: "runway.documents.list" }
   # list documents · read · 1 upstream call
   check { action: "runway.documents.delete", input: { … } }
   # free: inputs, permissions, cost and confirmations, nothing sent
   ```

Runway API at a glance

**Sign-in**

Bearer token. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead.

**You provide**

`api_key`

**Limits we pace to**

300 calls a minute, per connection

**API covered**

RunwayML API 2024-11-06`api.dev.runwayml.com`

**Built from**

Runway API's documentation, read 23 October 2026[docs.dev.runwayml.com](https://docs.dev.runwayml.com/openapi.json) [docs.dev.runwayml.com](https://docs.dev.runwayml.com/api/)

**Service file**

Version 1.0.0, signed (Ed25519)

**Cached reads**

18 of 22 reads, for up to 5 minutes; a write clears what it affects

**Shared entities**

2, so ask can line Runway API up with your other systems

**Data**

Processed and stored in Sydney, Australia. No US region yet

Capabilities

## What an AI agent can do in Runway API.

65 actions in 32 areas. Every one carries its effect, so a key can allow the reads and hold the rest.

- **22** read look, never change
- **36** write create and update
- **7** destructive wait for a confirmation

1. ### Recipes
   7
   0 read · 7 write · 0 destructive
   Localize an ad image · Create a marketing stock image · Create a multi-shot video · Create a product ad video
2. ### Voices
   6
   2 read · 3 write · 1 destructive
   Create a voice · List voices · Preview a voice · Get a voice
3. ### Avatars
   5
   2 read · 2 write · 1 destructive
   Create avatar · List avatars · Get avatar · Update avatar
4. ### Documents
   5
   2 read · 2 write · 1 destructive
   List documents · Get document · Delete document · Create document
5. ### Routers
   5
   2 read · 2 write · 1 destructive
   Create Model Router · List Model Routers · Retrieve Model Router · Update Model Router
6. ### Avatar conversations
   3
   2 read · 0 write · 1 destructive
   List conversations · Get conversation · Delete conversation
7. ### Realtime sessions
   3
   1 read · 1 write · 1 destructive
   Create realtime session · Get realtime session · Cancel realtime session
8. ### Routed generation
   3
   0 read · 3 write · 0 destructive
   Routed audio generation · Routed image generation · Routed video generation

**24 more areas · 28 actions**

- Audit logs **2**
- Tasks **2**
- Workflow invocations **2**
- Workflows **2**
- Avatar usage **1**
- Avatar videos **1**
- Character performance **1**
- Image to video **1**
- Image upscale **1**
- Organization **1**
- Organization usage **1**
- Router requests **1**
- Sound effect **1**
- Speech to speech **1**
- Text to image **1**
- Text to speech **1**
- Text to video **1**
- Uploads **1**
- Video to HDR **1**
- Video to video **1**
- Video upscale **1**
- Voice dubbing **1**
- Voice isolation **1**
- Workspace usage **1**

Top actions

## The Runway API API, shaped for agents.

The ids your agent types, what each does, and the one name it shares with other systems. 7 of Runway API's 7 action names are shared with at least one other connector, so an agent learns them once.

| Action | What it does | Effect | Same name in |
| --- | --- | --- | --- |
| `runway.documents.list` List documents | List documents. | read | `content.document.list` 28 connectors |
| `runway.workflows.list` List published workflows | List published workflows. | read | `automation.workflow.list` 51 connectors |
| `runway.documents.get` Get document | Get document. | read | `content.document.get` 34 connectors |
| `runway.workflows.get` Get workflow details | Get workflow details. | read | `automation.workflow.get` 51 connectors |
| `runway.audit_logs.list` List linked workspace audit logs | List linked workspace audit logs. | read | `identity.audit_event.list` 51 connectors |
| `runway.documents.delete` Delete document | Delete document. | destructive | `content.document.delete` 28 connectors |
| `runway.workflow_invocations.get` Get workflow invocation detail | Get workflow invocation detail. | read | `automation.workflow_run.get` 27 connectors |
| `runway.voices.list` List voices | List voices. | read | this connector |
| `runway.avatars.list` List avatars | List avatars. | read | this connector |
| `runway.voices.create` Create a voice | Create a voice. | write | this connector |
| `runway.avatars.create` Create avatar | Create avatar. | write | this connector |
| `runway.routers.create` Create Model Router | Create Model Router. | write | this connector |

Try it

## Prompts to paste into Claude.

Each one names the Runway API actions it will call, so you know what it reads and what it changes before you send it.

1. Read read
   List documents in Runway API and tell me what stands out.
   calls `runway.documents.list`
2. Write write
   Create a voice in Runway API. Show me exactly what you will send before you save it.
   calls `runway.voices.create`

Governance for Runway API

## Destructive actions wait for a yes.

7 Runway API actions are destructive. These are all of them:

- Delete document `runway.documents.delete`
- Cancel or delete a task `runway.tasks.delete`
- Delete a voice `runway.voices.delete`
- Delete avatar `runway.avatars.delete`
- Delete Model Router `runway.routers.delete`
- Delete conversation `runway.avatar_conversations.delete`
- Cancel realtime session `runway.realtime_sessions.cancel`

A destructive action runs only with `confirm: true`. Under the `human` policy the agent cannot confirm for itself: the run stops until a named person approves, in the console or from a signed link in an email. The cap on a key is refused before the call is sent, never after.

do runway.documents.delete confirm_policy: **human**

Delete document?

**intent:** the user asked me to delete document

run suspended · waiting for a named person · key _runway-agent_ cannot confirm itself

In the console, or from an emailed signed link.

a key for this connector

```
account {
  op: "keys.create", name: "runway-agent",
  allowed_actions: ["runway.*"], cap_calls: 2000,
  confirm_policy: "human_for_writes",
  approval_rules: [
    { match: "runway.documents.delete", approver: "lead@example.com" },
    { match: "runway.tasks.delete", approver: "lead@example.com" },
    { match: "runway.voices.delete", approver: "lead@example.com" }
  ]
}
```

Illustrative. Scoped to Runway API only, capped at 2,000 upstream calls, every write confirmed, and the destructive ones approved by a named person.

Cost

## What Runway API costs on actionlake.

Every Runway API action is one upstream call, the only thing actionlake bills. Searching, `check`, `cost`, `account` and any call Runway API refused with 401, 403, 429 or a 5xx are free. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.

- 1 upstream call per action all 65 Runway API actions
- 65 actions page each page of a long list is one call
- 18 reads can be cached a cache hit costs nothing
- 100,000 calls a month, free paid plans from US$49 a month for 250,000

[See pricing](https://actionlake.com/pricing)

Action index

## Every Runway API action.

65 actions by area: the id your agent calls, what it does, and its effect.

**Recipes 7**

- `runway.recipes.ad_localization` Localize an ad image write
- `runway.recipes.marketing_stock_image` Create a marketing stock image write
- `runway.recipes.multi_shot_video` Create a multi-shot video write
- `runway.recipes.product_ad` Create a product ad video write
- `runway.recipes.product_campaign_image` Create product campaign images write
- `runway.recipes.product_swap` Swap a product in a reference video write
- `runway.recipes.product_ugc` Create a product UGC video write

**Voices 6**

- `runway.voices.create` Create a voice write
- `runway.voices.delete` Delete a voice destructive
- `runway.voices.get` Get a voice read
- `runway.voices.list` List voices read
- `runway.voices.preview` Preview a voice write
- `runway.voices.update` Update a voice write

**Avatars 5**

- `runway.avatars.create` Create avatar write
- `runway.avatars.delete` Delete avatar destructive
- `runway.avatars.get` Get avatar read
- `runway.avatars.list` List avatars read
- `runway.avatars.update` Update avatar write

**Documents 5**

- `runway.documents.create` Create document write
- `runway.documents.delete` Delete document destructive
- `runway.documents.get` Get document read
- `runway.documents.list` List documents read
- `runway.documents.update` Update document write

**Routers 5**

- `runway.routers.create` Create Model Router write
- `runway.routers.delete` Delete Model Router destructive
- `runway.routers.get` Retrieve Model Router read
- `runway.routers.list` List Model Routers read
- `runway.routers.update` Update Model Router write

**Avatar conversations 3**

- `runway.avatar_conversations.delete` Delete conversation destructive
- `runway.avatar_conversations.get` Get conversation read
- `runway.avatar_conversations.list` List conversations read

**Realtime sessions 3**

- `runway.realtime_sessions.cancel` Cancel realtime session destructive
- `runway.realtime_sessions.create` Create realtime session write
- `runway.realtime_sessions.get` Get realtime session read

**Routed generation 3**

- `runway.routed_generation.audio` Routed audio generation write
- `runway.routed_generation.image` Routed image generation write
- `runway.routed_generation.video` Routed video generation write

**Audit logs 2**

- `runway.audit_logs.get` Get a linked workspace audit log entry read
- `runway.audit_logs.list` List linked workspace audit logs read

**Tasks 2**

- `runway.tasks.delete` Cancel or delete a task destructive
- `runway.tasks.get` Get task detail read

**Workflow invocations 2**

- `runway.workflow_invocations.create` Run a published workflow write
- `runway.workflow_invocations.get` Get workflow invocation detail read

**Workflows 2**

- `runway.workflows.get` Get workflow details read
- `runway.workflows.list` List published workflows read

**Avatar usage 1**

- `runway.avatar_usage.get` Get avatar usage read

**Avatar videos 1**

- `runway.avatar_videos.create` Generate avatar video from audio or text write

**Character performance 1**

- `runway.character_performance.create` Control a character write

**Image to video 1**

- `runway.image_to_video.create` Image to video write

**Image upscale 1**

- `runway.image_upscale.create` Image upscale write

**Organization 1**

- `runway.organization.get` Get organization information read

**Organization usage 1**

- `runway.organization_usage.query` Query credit usage read

**Router requests 1**

- `runway.router_requests.list` List Model Router requests read

**Sound effect 1**

- `runway.sound_effect.create` Generate sound effects write

**Speech to speech 1**

- `runway.speech_to_speech.create` Speech to speech write

**Text to image 1**

- `runway.text_to_image.create` Text/Image to Image write

**Text to speech 1**

- `runway.text_to_speech.create` Text to speech write

**Text to video 1**

- `runway.text_to_video.create` Text to video write

**Uploads 1**

- `runway.uploads.create` Upload a file write

**Video to HDR 1**

- `runway.video_to_hdr.create` Video to HDR write

**Video to video 1**

- `runway.video_to_video.create` Video to video write

**Video upscale 1**

- `runway.video_upscale.create` Video upscale write

**Voice dubbing 1**

- `runway.voice_dubbing.create` Voice dubbing write

**Voice isolation 1**

- `runway.voice_isolation.create` Voice isolation write

**Workspace usage 1**

- `runway.workspace_usage.list` List linked workspace usage read

Questions

## The Runway API MCP server, answered.

**Is there a Runway API MCP server for Claude?**

Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 65 Runway API actions across recipes, voices, avatars and documents, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by Runway API.

**What do I need to connect Runway API?**

Runway API secret key: in the Runway developer portal (https://dev.runwayml.com) open your organization > API Keys >…. The free call account { op: "connections.requirements", service: "runway" } lists exactly what to paste.

**Can an AI agent delete or change things in Runway API?**

Only if its key allows it, and only with an explicit confirmation. 7 Runway API actions are destructive, including delete document, cancel or delete a task and delete a voice. Each runs only with confirm: true. Under the human confirm policy, or an approval rule on the key, the run waits until a named person approves it.

**Will an AI agent use up my Runway API API limit?**

actionlake paces every call to the limits in the Runway API service file: 300 calls a minute per connection. Repeated reads are answered from cache (18 Runway API actions can be), which never reaches Runway API.

**What does a Runway API action cost?**

One upstream call, the only unit actionlake bills. A list that pages bills one call per page (65 Runway API actions page). Cache hits, searches, dry runs and calls Runway API refused with 401, 403, 429 or a 5xx are free. The Free plan includes 100,000 calls a month with no card. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.

**Which Runway API API does it cover?**

The RunwayML API 2024-11-06 at api.dev.runwayml.com. It is built from Runway API's documentation as read on 23 October 2026. 65 actions in 32 areas, each one in the signed service file (version 1.0.0).

**Does actionlake keep my Runway API data, and where?**

actionlake processes and stores data in Sydney, Australia; there is no US region yet. Reads may be cached for up to 5 minutes, and a write clears the reads it affects. Each run's request and response is kept, sealed, for your plan's trace window: 1 day on Free, 3 days on Starter, 7 days on Pro, 14 days on Team, 30 days on Business and 30 days on Enterprise. Your Runway API credential is sealed under your tenant's key, never logged and never returned. Your traffic is not used to train models.

**Which AI clients can use it?**

Claude Code, tested by us end to end, and the Claude app as a custom connector, tested by us with an existing key. Other MCP clients that support remote servers over Streamable HTTP (MCP revisions 2025-06-18 and 2025-11-25) and OAuth dynamic client registration, or a bearer header, should connect the same way; we have not tested them. Or plain REST, under the same action names.

## More connectors like Runway API

Shares the most action names with

- [Voiceflow 5](https://actionlake.com/connectors/voiceflow)
- [Cartesia 3](https://actionlake.com/connectors/cartesia)
- [Granola 1](https://actionlake.com/connectors/granola)
- [LangSmith 1](https://actionlake.com/connectors/langsmith)
- [OpenAI Admin 1](https://actionlake.com/connectors/openai-admin)

Joins with, on shared entities

- [ScalePad 1](https://actionlake.com/connectors/scalepad)
- [Vanta 1](https://actionlake.com/connectors/vanta)
- [Filevine 1](https://actionlake.com/connectors/filevine)
- [Clio Manage 1](https://actionlake.com/connectors/clio-manage)
- [NocoDB 1](https://actionlake.com/connectors/nocodb)
- [Basecamp 1](https://actionlake.com/connectors/basecamp)

Browse

- [All AI connectors](https://actionlake.com/connectors/category/ai)
- [All connectors](https://actionlake.com/connectors)
- [How actionlake compares](https://actionlake.com/compare)

actionlake is independent. It is not made or endorsed by Runway API; the name is the vendor's own. Counts are from the Runway API service file, version 1.0.0.

## Give your agent Runway API today.

[Connect Runway API free](https://actionlake.com/console?start=sandbox) [See pricing](https://actionlake.com/pricing)

Start free - 100,000 calls/month, no card. No signup. Hard caps on every key. Cached answers are free.

---

Source: https://actionlake.com/connectors/runway-api
More: https://actionlake.com/llms.txt