DeepAgents

Connect a DeepAgents agent to Civic using langchain-mcp-adapters, which bridges DeepAgents' tool interface with Civic's Streamable HTTP MCP transport.

Prerequisites

• Python 3.11+
• A Civic account at app.civic.com with a configured toolkit
• A Civic token (generate from Install → MCP URL)
• An API key for your chosen LLM provider (e.g., Anthropic, OpenAI)

Installation

Using uv (recommended):

uv add deepagents langchain-mcp-adapters langchain-anthropic fastapi uvicorn python-dotenv

Or with pip:

pip install deepagents langchain-mcp-adapters langchain-anthropic fastapi uvicorn python-dotenv

Environment Variables

# Civic token generated from app.civic.com → Install → MCP URL
CIVIC_TOKEN=your-civic-token

# API key for your chosen LLM provider
ANTHROPIC_API_KEY=your-anthropic-key   # or OPENAI_API_KEY, etc.

Get Your Credentials

How to generate a Civic token and configure toolkit URL parameters

Connecting to Civic

Use MultiServerMCPClient to connect to the Civic MCP Hub during app startup, then pass the discovered tools to create_deep_agent:

import os
from contextlib import asynccontextmanager
from fastapi import FastAPI
from deepagents import create_deep_agent
from langchain_mcp_adapters.client import MultiServerMCPClient

mcp_client = None
agent = None

@asynccontextmanager
async def lifespan(app: FastAPI):
    global mcp_client, agent

    mcp_client = MultiServerMCPClient({
        "civic-nexus": {
            "transport": "streamable_http",
            "url": "https://app.civic.com/hub/mcp",
            "headers": {"Authorization": f"Bearer {os.environ['CIVIC_TOKEN']}"},
        }
    })

    tools = await mcp_client.get_tools()

    agent = create_deep_agent(
        model="anthropic:claude-sonnet-4-6",  # or "openai:gpt-4o", etc.
        tools=tools,
        system_prompt="You are a helpful assistant with access to calendar and email tools.",
    )

    yield

    await mcp_client.__aexit__(None, None, None)

app = FastAPI(lifespan=lifespan)

Running the Agent

from fastapi import FastAPI
from pydantic import BaseModel

class ChatRequest(BaseModel):
    message: str
    thread_id: str = "default"

@app.post("/chat")
async def chat(request: ChatRequest):
    result = await agent.ainvoke(
        {"messages": [{"role": "user", "content": request.message}]},
        config={"configurable": {"thread_id": request.thread_id}},
    )
    return {"response": result["messages"][-1].content}

Start the server:

uv run uvicorn main:app --reload

Production Configuration

Lock to a Toolkit

For production agents, scope to a specific toolkit by adding profile to the MCP URL:

mcp_client = MultiServerMCPClient({
    "civic-nexus": {
        "transport": "streamable_http",
        "url": f"https://app.civic.com/hub/mcp?profile={os.environ['CIVIC_PROFILE_ID']}",
        "headers": {"Authorization": f"Bearer {os.environ['CIVIC_TOKEN']}"},
    }
})

When a profile is specified, the session is locked by default — the agent cannot switch toolkits or modify its own guardrails. This prevents prompt injection attacks from escaping the defined tool scope.

Environment Variable Reference

Variable	Description
CIVIC_TOKEN	Civic token from app.civic.com → Install → MCP URL
ANTHROPIC_API_KEY	API key for your LLM provider (e.g., Anthropic, OpenAI)

Reference Implementation

deepagents-reference-implementation-civic

Complete implementation with FastAPI server, streaming chat UI, and production patterns

Next Steps

Agent Deployment

Production deployment guide: profile locking, URL params, authentication

Guardrails

Constrain what tools your DeepAgents agent can use

Audit Trail

Query what your agent did via Civic Chat

Get Credentials

Token generation and URL parameter reference