Connect client

@10xscale/agentflow-client is a fully typed TypeScript client for every API endpoint exposed by agentflow api — graph execution, thread management, long-term memory, and file uploads.

Make sure the API server is running:

agentflow api --host 127.0.0.1 --port 8000

Install

npm install @10xscale/agentflow-client

Create a client

import { AgentFlowClient } from "@10xscale/agentflow-client";

const client = new AgentFlowClient({
  baseUrl: "http://127.0.0.1:8000",
});

If your server has auth enabled:

import { AgentFlowClient, bearerAuth } from "@10xscale/agentflow-client";

const client = new AgentFlowClient({
  baseUrl: "http://127.0.0.1:8000",
  auth: bearerAuth("your-api-token"),
});

Auth helpers: bearerAuth(token), basicAuth(username, password), headerAuth(name, value).

Make your first call

import { Message } from "@10xscale/agentflow-client";

const result = await client.invoke(
  [Message.text_message("What is the weather in London?")],
  {
    config: { thread_id: "my-thread-001" },
    recursion_limit: 10,
  }
);

console.log(result.messages.at(-1)?.text());

Stream responses

import { StreamEventType } from "@10xscale/agentflow-client";

const stream = client.stream(
  [Message.text_message("Tell me a long story.")],
  { config: { thread_id: "my-thread-002" } }
);

for await (const chunk of stream) {
  if (chunk.event === StreamEventType.MESSAGE && chunk.message) {
    process.stdout.write(chunk.message.text());
  }
}

Go deeper

The client covers three API layers — explore the how-to guides for full usage:

Topic	Guide
Client setup, auth, config options	Create a client
Invoke, stream, WebSocket, partial results	Invoke an agent
Streaming responses in depth	Stream responses
Thread state, messages, history	Manage threads
Long-term memory store and search	Use memory API
File uploads and multimodal messages	Upload files
Remote tools from the client side	Register remote tools

Next step

Run agentflow play to open the hosted playground and chat with your agent interactively.