Agent Update

Connect

LangGraph / LangChain → Agent Update

LangChain's Python stack for stateful, multi-step agents, wired to MCP through the langchain-mcp-adapters package.

LangGraph / LangChain talks to Agent Update over MCP. The server is https://api.tryagentupdate.com/v1/mcp, the transport is Streamable HTTP, and the credential is one header.

Setup

Python
# pip install langchain-mcp-adapters langchain
import os
from langchain_mcp_adapters.client import MultiServerMCPClient
from langchain.agents import create_agent

client = MultiServerMCPClient({
    "agent_update": {
        "transport": "streamable_http",
        "url": "https://api.tryagentupdate.com/v1/mcp",
        "headers": {"Authorization": f"Bearer {os.environ['AGENT_UPDATE_TOKEN']}"},
    }
})

tools = await client.get_tools()
agent = create_agent("anthropic:claude-opus-4-5", tools)

Before you file a bug

Use streamable_http, with the underscore. "streamable-http" and "http" work at runtime, but only the underscore form type-checks. Headers are fixed for the client's lifetime and there is no per-request header hook, so a rotating token needs an httpx.Auth passed as auth=. A static au_live_ PAT is fine. MultiServerMCPClient opens a fresh session per tool call; use async with client.session("agent_update") to hold one open. The JS package @langchain/mcp-adapters was **not verified**, so don't publish a JS form.

Check it worked

Ask the agent to text you. A 401 means the token is wrong, or the word Bearer is in there twice. No tool at all means the transport string is wrong. Both are covered in troubleshooting.

Next

The five tools and their arguments are on the MCP page. The same tools over plain HTTP are on the REST page. Every other tool is on the docs index.

← All docs