Create agent

Defines an agent in the caller's org: a model, a system prompt (instructions) and a set of tool names.

POST /v1/agent

Agent → · All API references → · Get started →


Address	`https://api.hanzo.ai/v1/agent`
Method	`POST`
Operation	`post_agent`
Auth	`Authorization: Bearer $HANZO_API_KEY`

Defines an agent in the caller's org: a model, a system prompt (instructions) and a set of tool names. The name must be unique in the org and match ^[A-Za-z0-9][A-Za-z0-9._-]{0,63}$. An omitted model takes the deployment's configured default; a named one is checked against the gateway's served catalog, so a model this deployment never serves is refused here rather than failing at run time. A long-running agent must carry a 5-field cron schedule (the scheduler would otherwise never fire it) and counts against a per-org cap on scheduled agents.

Request

14 fields, body application/json (required).

Field	In	Type	Required	Description
`avatar`	body	string	—	Avatar and Emoji are how the agent APPEARS.
`cap_micro_usd`	body	integer (int64)	—	The budget, required. An agent is not creatable without one: a zero cap would mean no limit and no record of its spend, which is what a budget exists to…
`computeRef`	body	string	—	ComputeRef optionally binds this bot to a visor machine.
`description`	body	string	—	Description is the one line published as the description of the `agent_<name>` tool, which is how another agent decides whether to call this one.
`emoji`	body	string	—	Emoji is the single glyph shown when there is no image.
`executionMode`	body	string	—	ExecutionMode is one-shot or long-running. Empty takes one-shot, which runs only when something POSTs to it.
`instructions`	body	string	—	Instructions is the system prompt, up to 32 KiB, stored verbatim.
`max_task_micro_usd`	body	integer (int64)	—	MaxTaskMicroUSD is the ceiling for a single run, in micro-USD.
`model`	body	string	—	Model names the model to run on. Omit it to take the deployment's configured default; name one and it is checked against the gateway's served catalogue here,…
`name`	body	string	—	Name is the agent's org-unique handle and the only required field.
`period`	body	string	—	Period is the window the cap resets on: day, week or month.
`schedule`	body	string	—	Schedule is the 5-field cron a long-running agent fires on, parsed here so a bad expression is a 400 and not an agent that silently never runs.
`serviceAccountId`	body	string	—	ServiceAccountID optionally names the IAM agent service account (<org>-<agent>) a scheduled run should be billed AS, so an autonomous run is attributable to a…
`tools`	body	string[]	—	Tools are the tool names this agent may call. Omitted or empty grants NONE — that default is the agent's authority and is not widened anywhere.

Response

Status	Body	Meaning
`201`	agent.agentView	created
`default`	problem-details	refused

201 body — 19 fields.

Field	In	Type	Always	Description
`avatar`	body	string	—	Avatar is an image the agent is drawn as — a link to one, or the bytes inline as a data URL, up to 96 KiB.
`cap_micro_usd`	body	integer (int64)	—	CapMicroUSD is the total this agent may spend within one Period, as an integer number of micro-USD (1,000,000 = $1).
`computeRef`	body	string	—	ComputeRef is the visor machine this bot is bound to, opaque here: this package stores and echoes it, and the binding's lifecycle belongs elsewhere.
`consumed_micro_usd`	body	integer (int64)	—	ConsumedMicroUSD is what has been spent in the current period, in micro-USD.
`createdAt`	body	string	—	CreatedAt is when the agent was defined, RFC 3339 in UTC to the second.
`description`	body	string	—	Description is the one line another agent reads when deciding whether to call this one: the tool catalogue publishes it as the description of `agent_<name>`,…
`emoji`	body	string	—	Emoji is the single glyph a caller picked when they had no image.
`executionMode`	body	string	—	ExecutionMode is one-shot or long-running, and it decides who may start this agent.
`id`	body	string	—	ID is the agent's stable handle, minted here as "agent_" + 32 hex characters of crypto/rand.
`max_task_micro_usd`	body	integer (int64)	—	MaxTaskMicroUSD is the ceiling for a single run, in micro-USD.
`model`	body	string	—	Model is the Zen model this agent runs on, and it is always OUR name for it: writes normalize through cloud.ZenModel and the read normalizes again, so an…
`name`	body	string	—	Name is the agent's org-unique handle, matching ^[A-Za-z0-9][A-Za-z0-9._-]{0,63}$.
`period`	body	string	—	Period is the window the cap resets on: day, week or month.
`runs`	body	integer (int64)	—	Runs is how many executions the org has recorded against this agent, counted at read time.
`schedule`	body	string	—	Schedule is the 5-field cron the scheduler fires a long-running agent on, evaluated once a minute.
`serviceAccountId`	body	string	—	ServiceAccountID is the IAM agent service account (<org>-<agent>) a scheduled run is billed AS.
`status`	body	string	—	Status is the agent's readiness, and today it is "ready" on every row: an agent is a definition rather than a provisioned thing, so nothing transitions it.
`tools`	body	string[]	—	Tools are the tool names this agent may call, and the list IS the authority: an agent that declares none gets none.
`updatedAt`	body	string	—	UpdatedAt is the last time any field above was written, same format.

Failure carries the platform error shape — see Errors.

Examples

hanzo agent create

import { Configuration, AgentApi } from 'hanzoai';

const api = new AgentApi(new Configuration({ accessToken: process.env.HANZO_API_KEY }));
const { data } = await api.postAgent({ avatar: "<avatar>", cap_micro_usd: 0 });

from hanzoai.cloud import ApiClient, Configuration
from hanzoai.cloud.api import AgentApi

client = ApiClient(Configuration(access_token=os.environ["HANZO_API_KEY"]))
result = AgentApi(client).post_agent(avatar="<avatar>", cap_micro_usd=0)

cfg := hanzoai.NewConfiguration()
cfg.AddDefaultHeader("Authorization", "Bearer "+os.Getenv("HANZO_API_KEY"))
client := hanzoai.NewAPIClient(cfg)

resp, _, err := client.AgentAPI.PostAgent(context.Background()).Execute()
if err != nil {
	return err
}

use hanzo_client::apis::{configuration::Configuration, agent_api};

let mut cfg = Configuration::new();
cfg.bearer_access_token = std::env::var("HANZO_API_KEY").ok();

let result = agent_api::post_agent(&cfg, Default::default()).await?;

import ai.hanzo.cloud.ApiClient;
import ai.hanzo.cloud.api.AgentApi;

ApiClient client = new ApiClient();
client.setBearerToken(System.getenv("HANZO_API_KEY"));

var result = new AgentApi(client).postAgent();

The method above is the one at the current release of the document. [email protected] (npm) was generated from an earlier release, where this operation carried a different id, so it spells the method differently — regenerating the clients is what makes the two agree. SDKs →

curl -X POST https://api.hanzo.ai/v1/agent \
  -H "Authorization: Bearer $HANZO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
       "avatar": "<avatar>",
       "cap_micro_usd": 0
     }'

Tool agents, op post_agent — POST the JSON-RPC envelope to https://api.hanzo.ai/v1/mcp.

curl -X POST https://api.hanzo.ai/v1/mcp \
  -H "Authorization: Bearer $HANZO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
       "jsonrpc": "2.0",
       "id": 1,
       "method": "tools/call",
       "params": {
         "name": "agents",
         "arguments": {
           "op": "post_agent",
           "input": {
             "avatar": "<avatar>",
             "cap_micro_usd": 0
           }
         }
       }
     }'

Agent API · All Hanzo APIs · Interactive reference

Request

Response

Examples

On this page