<!-- Generated from the canonical Teloring documentation source. Do not edit this file. -->

Canonical page: https://docs.teloring.com/api/create-ai-agent
Last modified: 2026-08-20T20:52:05.000Z

# Create an AI agent

import MethodEndpoint from "@theme/ApiExplorer/MethodEndpoint";
import ParamsDetails from "@theme/ParamsDetails";
import RequestSchema from "@theme/RequestSchema";
import StatusCodes from "@theme/StatusCodes";
import OperationTabs from "@theme/OperationTabs";
import TabItem from "@theme/TabItem";
import Heading from "@theme/Heading";
import Translate from "@docusaurus/Translate";

<Heading
  as={"h1"}
  className={"openapi__heading"}
  children={"Create an AI agent"}
>
</Heading>

<MethodEndpoint
  method={"post"}
  path={"/ai-agents"}
  context={"endpoint"}
>
  
</MethodEndpoint>



Create an AI agent. ⚠️ **This may charge the card on file** — an AI agent costs
a seat exactly as a human one does, with the same `402` /
`code: seat_charge_failed` behaviour and the same rollback.

Every field the console's AI agent editor offers is accepted and passed through:
persona, instructions, model, language, the knowledge bases it may answer from,
the inboxes it works in and its hand-off rules. Unknown fields are rejected
rather than silently dropped.

<Heading
  id={"request"}
  as={"h2"}
  className={"openapi-tabs__heading"}
>
  <Translate id="theme.openapi.request.title">Request</Translate>
</Heading>

<ParamsDetails>
  
</ParamsDetails>

<RequestSchema
  {...require("./create-ai-agent.RequestSchema.json")}
>
  
</RequestSchema>

<StatusCodes
  {...require("./create-ai-agent.StatusCodes.json")}
>
  
</StatusCodes>
