API reference - Onecortex

The public API is one operation: invoke an agent. Everything else, creating agents, deploying, configuration and keys, is done in the dashboard. The dashboard’s own API is not a public interface, and may change without notice.

The playground on the operation’s page calls https://api.onecortex.io with your key: it answers once your organization is live there. During the private beta, use the dashboard’s chat console or curl to try an agent.

Base URL

https://api.onecortex.io

HTTPS only.

Authentication

Every call carries an API key as a bearer token:

Authorization: Bearer $ONECORTEX_API_KEY

Create a key under API keys in the dashboard. It is shown once, when you create it. See API keys.

Errors

A request refused before the run starts answers with an HTTP status and this body:

{
  "error": {
    "code": "not_found",
    "message": "That agent does not exist.",
    "requestId": "req_01K..."
  }
}

code is stable, and safe to branch on. message is for people, and may change. Every code, its status and what to do is on Errors.

Rate limit headers

Header	Holds
`X-RateLimit-Remaining`	Requests left in your organization’s burst allowance
`Retry-After`	On `429`: whole seconds to wait

The limits are on Limits.

The operation

POST /v1/agents/{agentId}/invoke: the request, both response shapes and every event are on its page in this tab, and explained with examples on the invoke API.

Last modified on September 28, 2026

​Base URL

​Authentication

​Errors

​Rate limit headers

​The operation

Base URL

Authentication

Errors

Rate limit headers

The operation