# apifare for LangGraph

One prepaid, capped key for every API your agent needs, under hard caps and a kill switch.

1. Get a token: https://apifare.com/start?origin=page:integrations-langgraph (shown once).
2. In LangGraph v1, the prebuilt agent is create_agent from langchain.agents, which runs on LangGraph and replaces create_react_agent. Load apifare's tools with LangChain's MCPAdapter and pass them in: the agent you get back is a compiled LangGraph graph.

```
pip install "langchain[mcp]"
```

```
import asyncio, os
from fastmcp.client import Client
from langchain.agents import create_agent
from langchain.mcp import MCPAdapter

async def main():
    client = Client("https://apifare.com/mcp", auth=os.environ["APIPAY_TOKEN"])
    async with MCPAdapter(client) as adapter:
        tools = await adapter.list_tools()
        agent = create_agent("claude-sonnet-5", tools)
        result = await agent.ainvoke(
            {"messages": [{"role": "user", "content": "What's the weather in London?"}]}
        )
        print(result["messages"][-1].content)

asyncio.run(main())
```

   Set APIPAY_TOKEN and your model provider's key, then run it. The token goes in as auth without the word "Bearer": FastMCP adds it.
3. Try it: "search the web for X", then "fetch a URL", then "what's the weather in London".
4. Caps and the kill switch are checked before dispatch. An empty balance returns HTTP 402 with a top-up link (see /errors.md).

Prices: https://apifare.com/capabilities. Config checked against https://docs.langchain.com/oss/python/migrate/langgraph-v1 on 2026-09-13.