Send every agent run to your own systems with a webhook

Add a webhook to an agent and Orkeste posts the result of each run to the URL you choose, as JSON. To start agents from your systems, use the Agent API.

What your agents do in Webhooks

A webhook sends results out; it does not let the agent call arbitrary APIs during a run.

Change

Sent after each run, without an approval step.

3 ways to use Webhooks with your agents

Results into your app

Each run's output lands in your backend, where you store it or show it to your users.

Hand-off to other automations

Point the webhook at a tool you already use for automation, and continue the process there.

Two-way with the Agent API

Start a run from your system through the API, and receive the result by webhook.

By team

Getting started

  1. Connect Webhooks: Grant access from the Connectors page. No technical setup.
  2. Describe the agent's job: In plain language, or from one of the 29 templates.
  3. Approve what matters: Actions you mark as sensitive wait in Validations until you approve them.

Security and control

Frequently asked questions

Can a webhook trigger an agent?

Yes, with an event trigger: in Agent Studio, add a webhook trigger to the agent and post to the URL it gives you, signed with HMAC or with the secret as a Bearer token. The webhook connector on this page is the other direction: it sends each run's result out.

Do I need coding skills to connect Webhooks?

No. You connect Webhooks from the Connectors page by granting access, then attach it to the agents that need it.

Is my Webhooks data protected?

Yes. Access goes through the service authorization, tokens are encrypted with AES-256 and isolated per workspace, and data is encrypted in transit and at rest.

Can I remove access at any time?

Yes. Disconnect Webhooks from the Connectors page and your agents lose access.

Often used with

Keep reading