AI and media

AI agents that use Runway

UtilitySpaces gives an AI agent scoped access to Runway through its own encrypted connection, not your password. The agent can call 3 named Runway tools, and the 2 that write wait for a person's approval unless an admin has set the connection to run without asking. Every call is written to a hash-chained audit log with its arguments and its result.

What an agent can do in Runway

Generate video from an image with Runway Gen-3 and track the resulting tasks. An agent calls these by name, and matching is exact.

Every tool a Runway connection exposes, and whether it writes, which waits for approval by default
Tool	What it does	Waits for approval by default
Image to video	Start an image-to-video generation (billed run). Returns a task id — poll get_task until the task succeeds or fails.	Yes — it writes
Get task	Get a generation task's status and output URLs.	No — read only
Cancel task	Cancel a running task or delete a finished one.	Yes — it writes

1 read tool, 2 write tools. Generated from the connector catalogue the product runs on, so it cannot list a tool that does not exist.

What teams hand an agent in Runway

Video teams hand an agent the render pipeline around Runway: Image to video starts a Gen-3 generation from a still, Get task reports how the render is going, and Cancel task stops one. Starting a render is a billed write that waits for a person unless this connection is set to run without asking; Cancel task is a delete, and a delete always waits for a person, whatever the policy.

Start an image-to-video render and post the output URL when the task succeeds (billed; after a person approves it)
Report the status of the renders in flight when the team asks what is still processing.
Flag a run that was started with the wrong source so a person can cancel it in Runway before it finishes billing

Before you let it write, or keep out: Renders cost money the moment they start. Think before letting Image to video run without asking, and have the agent quote the source image and prompt in the thread so the team knows exactly what is being rendered.

What waits for a person

Reads run on their own. These do not — each one waits in Approvals for an owner or admin unless an admin has set this connection to run writes without asking, and every call toRunway, approved, rejected or unattended, is logged.

Image to video POST
Cancel task DELETE

What connecting Runway takes

Runway uses an API key you create in Runway and paste in once. A workspace admin sets the connection up; the fields are:

API Secret

Credentials are encrypted and bound to your workspace, never read back, and every outbound call passes one guarded fetch — the connector security posture sets out exactly how, and the security page covers the platform around it.

Where Runway fits

Give an agent Runway work by writing a message in the inbox, or put its tools in a scheduled flow. What the run cost is on the cost breakdown.

Other ai and media integrations: AnythingLLM, AssemblyAI, D-ID. Or see every application agents can connect to.

See it doing your actual work

Open a demo workspace, connect one app, and give an agent a real job. No card, no sales call.

Try the demo See pricing

We reply to every message within one business day.