POST
/
v1
/
agent-tasks
Create a task
curl --request POST \
  --url https://platform.jobbit.uk/v1/agent-tasks \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "goal": "Write a headline and meta description for the Kent Physio home page.",
  "input": {
    "site": "https://kentphysio.co.uk",
    "tone": "warm and plain"
  },
  "output_schema": {
    "type": "object",
    "properties": {
      "headline": {
        "type": "string"
      },
      "meta_description": {
        "type": "string",
        "maxLength": 160
      }
    },
    "required": [
      "headline",
      "meta_description"
    ]
  },
  "workspace": {
    "key": "kent-physio-site",
    "title": "Kent Physio website"
  },
  "model": "jobbit/pro",
  "effort": "high",
  "limits": {
    "max_credits": 200,
    "max_minutes": 20
  },
  "on_question": "fail",
  "idempotency_key": "kp-home-copy-2026-11-05",
  "source": {
    "ticket": "KP-142"
  }
}
'
import requests

url = "https://platform.jobbit.uk/v1/agent-tasks"

payload = {
    "goal": "Write a headline and meta description for the Kent Physio home page.",
    "input": {
        "site": "https://kentphysio.co.uk",
        "tone": "warm and plain"
    },
    "output_schema": {
        "type": "object",
        "properties": {
            "headline": { "type": "string" },
            "meta_description": {
                "type": "string",
                "maxLength": 160
            }
        },
        "required": ["headline", "meta_description"]
    },
    "workspace": {
        "key": "kent-physio-site",
        "title": "Kent Physio website"
    },
    "model": "jobbit/pro",
    "effort": "high",
    "limits": {
        "max_credits": 200,
        "max_minutes": 20
    },
    "on_question": "fail",
    "idempotency_key": "kp-home-copy-2026-11-05",
    "source": { "ticket": "KP-142" }
}
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}

response = requests.post(url, json=payload, headers=headers)

print(response.text)
const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: JSON.stringify({
    goal: 'Write a headline and meta description for the Kent Physio home page.',
    input: {site: 'https://kentphysio.co.uk', tone: 'warm and plain'},
    output_schema: {
      type: 'object',
      properties: {headline: {type: 'string'}, meta_description: {type: 'string', maxLength: 160}},
      required: ['headline', 'meta_description']
    },
    workspace: {key: 'kent-physio-site', title: 'Kent Physio website'},
    model: 'jobbit/pro',
    effort: 'high',
    limits: {max_credits: 200, max_minutes: 20},
    on_question: 'fail',
    idempotency_key: 'kp-home-copy-2026-11-05',
    source: {ticket: 'KP-142'}
  })
};

fetch('https://platform.jobbit.uk/v1/agent-tasks', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));
<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://platform.jobbit.uk/v1/agent-tasks",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "POST",
  CURLOPT_POSTFIELDS => json_encode([
    'goal' => 'Write a headline and meta description for the Kent Physio home page.',
    'input' => [
        'site' => 'https://kentphysio.co.uk',
        'tone' => 'warm and plain'
    ],
    'output_schema' => [
        'type' => 'object',
        'properties' => [
                'headline' => [
                                'type' => 'string'
                ],
                'meta_description' => [
                                'type' => 'string',
                                'maxLength' => 160
                ]
        ],
        'required' => [
                'headline',
                'meta_description'
        ]
    ],
    'workspace' => [
        'key' => 'kent-physio-site',
        'title' => 'Kent Physio website'
    ],
    'model' => 'jobbit/pro',
    'effort' => 'high',
    'limits' => [
        'max_credits' => 200,
        'max_minutes' => 20
    ],
    'on_question' => 'fail',
    'idempotency_key' => 'kp-home-copy-2026-11-05',
    'source' => [
        'ticket' => 'KP-142'
    ]
  ]),
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer <token>",
    "Content-Type: application/json"
  ],
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}
package main

import (
	"fmt"
	"strings"
	"net/http"
	"io"
)

func main() {

	url := "https://platform.jobbit.uk/v1/agent-tasks"

	payload := strings.NewReader("{\n  \"goal\": \"Write a headline and meta description for the Kent Physio home page.\",\n  \"input\": {\n    \"site\": \"https://kentphysio.co.uk\",\n    \"tone\": \"warm and plain\"\n  },\n  \"output_schema\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"headline\": {\n        \"type\": \"string\"\n      },\n      \"meta_description\": {\n        \"type\": \"string\",\n        \"maxLength\": 160\n      }\n    },\n    \"required\": [\n      \"headline\",\n      \"meta_description\"\n    ]\n  },\n  \"workspace\": {\n    \"key\": \"kent-physio-site\",\n    \"title\": \"Kent Physio website\"\n  },\n  \"model\": \"jobbit/pro\",\n  \"effort\": \"high\",\n  \"limits\": {\n    \"max_credits\": 200,\n    \"max_minutes\": 20\n  },\n  \"on_question\": \"fail\",\n  \"idempotency_key\": \"kp-home-copy-2026-11-05\",\n  \"source\": {\n    \"ticket\": \"KP-142\"\n  }\n}")

	req, _ := http.NewRequest("POST", url, payload)

	req.Header.Add("Authorization", "Bearer <token>")
	req.Header.Add("Content-Type", "application/json")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}
HttpResponse<String> response = Unirest.post("https://platform.jobbit.uk/v1/agent-tasks")
  .header("Authorization", "Bearer <token>")
  .header("Content-Type", "application/json")
  .body("{\n  \"goal\": \"Write a headline and meta description for the Kent Physio home page.\",\n  \"input\": {\n    \"site\": \"https://kentphysio.co.uk\",\n    \"tone\": \"warm and plain\"\n  },\n  \"output_schema\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"headline\": {\n        \"type\": \"string\"\n      },\n      \"meta_description\": {\n        \"type\": \"string\",\n        \"maxLength\": 160\n      }\n    },\n    \"required\": [\n      \"headline\",\n      \"meta_description\"\n    ]\n  },\n  \"workspace\": {\n    \"key\": \"kent-physio-site\",\n    \"title\": \"Kent Physio website\"\n  },\n  \"model\": \"jobbit/pro\",\n  \"effort\": \"high\",\n  \"limits\": {\n    \"max_credits\": 200,\n    \"max_minutes\": 20\n  },\n  \"on_question\": \"fail\",\n  \"idempotency_key\": \"kp-home-copy-2026-11-05\",\n  \"source\": {\n    \"ticket\": \"KP-142\"\n  }\n}")
  .asString();
require 'uri'
require 'net/http'

url = URI("https://platform.jobbit.uk/v1/agent-tasks")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Post.new(url)
request["Authorization"] = 'Bearer <token>'
request["Content-Type"] = 'application/json'
request.body = "{\n  \"goal\": \"Write a headline and meta description for the Kent Physio home page.\",\n  \"input\": {\n    \"site\": \"https://kentphysio.co.uk\",\n    \"tone\": \"warm and plain\"\n  },\n  \"output_schema\": {\n    \"type\": \"object\",\n    \"properties\": {\n      \"headline\": {\n        \"type\": \"string\"\n      },\n      \"meta_description\": {\n        \"type\": \"string\",\n        \"maxLength\": 160\n      }\n    },\n    \"required\": [\n      \"headline\",\n      \"meta_description\"\n    ]\n  },\n  \"workspace\": {\n    \"key\": \"kent-physio-site\",\n    \"title\": \"Kent Physio website\"\n  },\n  \"model\": \"jobbit/pro\",\n  \"effort\": \"high\",\n  \"limits\": {\n    \"max_credits\": 200,\n    \"max_minutes\": 20\n  },\n  \"on_question\": \"fail\",\n  \"idempotency_key\": \"kp-home-copy-2026-11-05\",\n  \"source\": {\n    \"ticket\": \"KP-142\"\n  }\n}"

response = http.request(request)
puts response.read_body
{
  "task_id": "c41d7e0a-5b92-4f63-8e1a-3a9d2f6b7c10",
  "status": "queued",
  "reason": null,
  "goal_completed": null,
  "output": null,
  "text": null,
  "artifacts": [],
  "files": [],
  "error": null,
  "metrics": {
    "cost_credits": null,
    "cost_usd": null,
    "turns": null,
    "duration_seconds": null
  },
  "chat_id": "9a2f6d1e-83c4-4b05-a7e9-0c5d1b4f2e68",
  "workspace": "kent-physio-site",
  "model": "jobbit/pro",
  "effort": "high",
  "on_question": "fail",
  "question": null,
  "questions_asked": 0,
  "budget_credits": 200,
  "source": {
    "ticket": "KP-142"
  },
  "created_at": "2026-11-05T14:02:11+00:00",
  "started_at": null,
  "finished_at": null
}
{
  "error": {
    "code": "invalid_key",
    "message": "the API key is missing or invalid",
    "request_id": "5d3f0b7c8a2e4e19b6c47a90d1e2f3a4"
  }
}
{
  "error": {
    "code": "insufficient_credits",
    "message": "not enough credits to run this request",
    "request_id": "5d3f0b7c8a2e4e19b6c47a90d1e2f3a4"
  }
}
{
  "error": {
    "code": "insufficient_scope",
    "message": "this key does not have the deploy:write permission",
    "details": {
      "required": "deploy:write"
    },
    "request_id": "5d3f0b7c8a2e4e19b6c47a90d1e2f3a4"
  }
}
{
  "error": {
    "code": "file_not_found",
    "message": "file e2f0a9b4-6d17-4c3a-95be-1f8a7d3c6420 not found",
    "request_id": "5d3f0b7c8a2e4e19b6c47a90d1e2f3a4"
  }
}
{
  "error": {
    "code": "idempotency_conflict",
    "message": "this idempotency_key was already used with a different request",
    "request_id": "5d3f0b7c8a2e4e19b6c47a90d1e2f3a4"
  }
}
{
  "error": {
    "code": "input_too_large",
    "message": "input is larger than 262144 bytes",
    "request_id": "5d3f0b7c8a2e4e19b6c47a90d1e2f3a4"
  }
}
{
  "error": {
    "code": "invalid_callback",
    "message": "callback url must use https",
    "request_id": "5d3f0b7c8a2e4e19b6c47a90d1e2f3a4"
  }
}
{
  "error": {
    "code": "rate_limited",
    "message": "too many read requests, retry in 4 s",
    "details": {
      "group": "read",
      "limit": 120,
      "window_seconds": 60
    },
    "request_id": "5d3f0b7c8a2e4e19b6c47a90d1e2f3a4"
  }
}

Authorizations

​
Authorization
string
header
required

An API key from Settings → API keys, sent as Authorization: Bearer jb_live_…

Body

application/json
​
goal
string
required

What to do and when the job counts as done.

Required string length: 1 - 4000
​
input
any

Data for the task, up to 256 KiB of JSON. The agent treats it as data, not as instructions.

​
files
string[]

Ids of files from POST /v1/files, up to 20. The key also needs files:read.

Maximum array length: 20
Required string length: 1 - 64
​
output_schema
object

A JSON Schema (draft 2020-12, up to 64 KiB) the result must follow. The task fails with invalid_output if the agent cannot meet it.

​
project_id
string

A project to run the task in. You need to be an editor of it.

Maximum string length: 64
​
workspace
object
​
model
string

A model id, such as jobbit/pro. Defaults to the default for the account's plan.

Maximum string length: 100
​
effort
enum<string>
default:high

How hard the agent thinks.

Available options:
none,
low,
medium,
high,
xhigh,
max
​
limits
object
​
on_question
enum<string>
default:fail

fail: the agent decides for itself. ask: it may stop with a question and wait in waiting_input.

Available options:
fail,
ask
​
callback
object
​
idempotency_key
string

Repeating a create with the same key and the same body returns the first task; the same key with a different body is a conflict.

Required string length: 1 - 200
​
source
object

Your own labels, up to 4 KiB. They are stored on the task and returned unchanged.

Response

The task was created. A repeated idempotency_key returns the existing task with status 200 instead.

​
task_id
string
required

The task id.

​
status
enum<string>
required

Where the task is.

Available options:
queued,
running,
waiting_input,
succeeded,
failed,
cancelled
​
reason
string | null

Why the task is in this state: completed, cancelled, question, invalid_output, budget_exceeded, timed_out, goal_unreachable, worker_stopped, insufficient_credits or run_failed.

​
goal_completed
boolean | null

Whether the agent reports that the goal was met.

​
output
any

The result following your output_schema. Only set when the task succeeded and a schema was given.

​
text
string | null

The agent's final answer as text.

​
artifacts
object[]

Files the agent made.

​
files
string[]

Ids of the files handed to the task.

​
error
string | null

What went wrong, for a failed task.

​
metrics
object
​
chat_id
string | null

The id of the chat in the Jobbit app where the work happens.

​
workspace
string | null

The workspace key, if one was given.

​
model
string

The model the task runs on.

​
effort
string

The effort level.

​
on_question
enum<string>

Whether the task may ask questions.

Available options:
fail,
ask
​
question
object | null

The open question, while the status is waiting_input.

​
questions_asked
integer

How many questions the task has asked.

​
budget_credits
number

The most credits the task may spend.

​
source
object

Your source labels.

​
created_at
string<date-time>

When the task was created.

​
started_at
string<date-time> | null

When it started running.

​
finished_at
string<date-time> | null

When it finished.