Configure specialized agents and connect them to internal/external MCP servers.
Step 2: Launch Fleet
Make a single API call to /chat/stream, spawn 1000+ subagents across multiple machines, and complete any task within minutes.
Step 3: Get Results
Get logs for each session, search and filter through thousands of activities and debug any issues.
Save months of dev time: no plumbing, no chat history or tool-calling or state to manage. Built-in memory, logs, deterministic guardrails, replay & monitoring, plus a free testing lab.
We scale your agents horizontally
We deploy your agents across multiple machines and make them work seamlessly with each other.
You
API
/chat/stream
machine 1us-east-1
machine 2us-west-2
machine 3us-east-2
"Research about transformers."
Stop Building Boilerplate. Start Building Products.
import json
import requests
url = "https://api.clonejutsu.dev/chat/stream"
headers = {
"Authorization": "Bearer YOUR_API_TOKEN",
"Content-Type": "application/json",
}
payload = {
"prompt": "Summarize the attached PDF",
"coordinator_agent_id": "00000000-0000-0000-0000-0000000000CC",
"session_id": "12345678-1234-1234-1234-1234567890AB",
}
with requests.post(url, headers=headers, data=json.dumps(payload), stream=True) as r:
for line in r.iter_lines():
if line:
print(line.decode())