Skip to content
Auteryn

Run an Agent

POST /api/run

The primary endpoint to send a message to your agent and receive a response.

Request

POST https://api.auteryn.ai/api/run

Body:

{
  "query": "string (required)",
  "task_id": "string (optional — continue an existing conversation)",
  "stream": true,
  "context": {
    "user_id": "string (optional)",
    "user_name": "string (optional)",
    "metadata": {}
  }
}
FieldRequiredDescription
query✅The message or task to send to the agent
task_id❌Pass to continue an existing conversation. Omit to start new.
stream❌true for SSE streaming (default), false for single JSON response
context.user_id❌Your internal user ID
context.metadata❌Any key-value pairs your instructions can reference

Responses

Content-Type: text/event-stream

data: {"type":"run_started","run_id":"run_abc","task_id":"task_xyz"}


data: {"type":"text_chunk","content":"Let me check the open bugs..."}


data: {"type":"tool_call","tool":"github","action":"list_issues","params":{"state":"open","label":"P1"}}


data: {"type":"tool_result","tool":"github","result":{"count":3}}


data: {"type":"text_chunk","content":"I found 3 open P1 issues:"}


data: {"type":"run_completed","run_id":"run_abc","credits_used":6,"duration_ms":3241}

Code examples

import httpx, json


with httpx.stream(
    "POST",
    "https://api.auteryn.ai/api/run",
    headers={
        "Authorization": "Bearer afk_your_key",
        "X-Agent-Id": "agent_xyz",
    },
    json={"query": "What are the open P1 GitHub issues?"},
) as r:
    for line in r.iter_lines():
        if line.startswith("data: "):
            event = json.loads(line[6:])
            if event["type"] == "text_chunk":
                print(event["content"], end="", flush=True)