How to Use the GrowthBook MCP in LangChain

Q: Can I trace GrowthBook flag updates within LangChain runs using LangSmith?

Yes, every time your agent calls updatefeature or togglefeature within a chain, LangSmith captures the complete payload and latency. This gives you a clear audit trail of which LLM run triggered a specific flag change.

Q: How does LangChain handle environment errors in GrowthBook?

If a tool like createenvironment fails due to a naming conflict, the LangChain agent catches the API error within the chain. The agent can then analyze the error message and automatically try again with a corrected payload.

Q: Can the agent manage multiple GrowthBook projects at once?

Yes, the agent can use listprojects to loop through your workspaces. It can then apply updates across different projects sequentially within a single run.

Chain GrowthBook flag toggles directly into your LangChain pipelines to automate staging setups and production rollouts.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect GrowthBook MCP to LangChain

Create your Vinkius account to connect GrowthBook to LangChain and route execution through our secure gateway. The platform manages server hosting, runtime updates, and security layers. Configuration requires no manual server provisioning.

GDPR Free for Subscribers

Setup GrowthBook with LangChain

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Automate flag creation across LangChain runs

The `create_feature` tool lets your LangChain agent spin up new flag configurations on the fly based on upstream chain outputs. When a new product requirement is parsed by your pipeline, the agent immediately writes the flag rules to GrowthBook without manual intervention. You can feed these new flag schemas directly into `create_environment` to ensure your staging setup matches production instantly. LangSmith traces every single step of this chain, so you see exactly when and why a flag was created during execution.

Chain MCP Server toggles with live validation

The `toggle_feature` tool gives your LangChain agent the power to turn features on or off during automated integration tests. Instead of writing custom API scripts, the agent runs your test suite, checks for failures, and flips the toggle state based on the test results. After toggling, the agent calls `get_feature` to verify the state change propagated correctly before moving to the next link in your deployment chain. This makes your release pipeline completely autonomous and self-healing.

Keep multi-project environments in sync

The `list_projects` tool allows your LangChain pipeline to discover all active workspaces and map them to corresponding code repositories. Your agent can query multiple projects, identify outdated flags, and clean them up systematically. By combining `list_features` with `delete_feature` in a single chain, you can purge dead code flags across your entire organization via the MCP server. The agent handles the tedious cleanup work while you monitor the execution path in your LangSmith dashboard.

Setup guide

Set up GrowthBook MCP in LangChain

Prerequisites

Python 3.10+ installed
langchain-mcp-adapters + langgraph packages
Active Vinkius subscription with a valid endpoint token

1

Install dependencies
Run pip install langchain-mcp-adapters langgraph langchain-openai. The MCP adapters package converts MCP tools into native LangChain BaseTool objects.
2

Connect via HTTP transport
Use MultiServerMCPClient with "transport": "http" pointing to your Vinkius endpoint. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.
3

Create a ReAct agent
Pass the discovered tools to create_react_agent() from LangGraph. The agent automatically routes GrowthBook tool calls through the MCP protocol.
4

Run with any LLM
Swap ChatOpenAI for ChatAnthropic, ChatGoogleGenerativeAI, or any LangChain-compatible model. The MCP tools work identically across all providers.

agent.py

from langchain_mcp_adapters.client import MultiServerMCPClient
from langgraph.prebuilt import create_react_agent
from langchain_openai import ChatOpenAI

async with MultiServerMCPClient({
    "growthbook-mcp": {
        "transport": "http",
        "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp",
    }
}) as client:
    tools = client.get_tools()

    agent = create_react_agent(
        ChatOpenAI(model="gpt-4o"),
        tools,
    )
    result = await agent.ainvoke({
        "messages": "List recent GrowthBook transactions"
    })
    print(result["messages"][-1].content)

Get your connection token →

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by GrowthBook. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.

Why Choose Vinkius

Vinkius connects your tools to AI with real-time monitoring and automatic cost savings — all from one dashboard.

Connect GrowthBook now

Real-time monitoring

Live

visibility into every interaction

Connect your favorite tools to your AI and see exactly what's happening — every request, every response, in real time.

Built-in savings

60%

lower AI costs

Vinkius compresses data between your apps and your AI automatically. Lower bills every month — no configuration required.

Single dashboard

One

place for every integration

Every tool your AI connects to, managed from a single screen. One account, complete control.

Common questions about GrowthBook MCP in LangChain

Install the langchain-mcp-adapters package and initialize the MultiServerMCPClient with the GrowthBook endpoint. Call client.get_tools() to retrieve the 15 feature and environment tools, then pass them directly into your agent constructor.

Yes, every time your agent calls `update_feature` or `toggle_feature` within a chain, LangSmith captures the complete payload and latency. This gives you a clear audit trail of which LLM run triggered a specific flag change.

If a tool like `create_environment` fails due to a naming conflict, the LangChain agent catches the API error within the chain. The agent can then analyze the error message and automatically try again with a corrected payload.

Yes, the agent can use `list_projects` to loop through your workspaces. It can then apply updates across different projects sequentially within a single run.

No, your GrowthBook API keys never touch the LangChain framework directly. Vinkius hosts the MCP server in a secure sandbox, meaning only the raw tool outputs like feature lists are returned to your local runtime.

Use it with your favorite AI tools

Connect this server to Cursor, Claude, VS Code, and more.

OpenAI Agents SDK sdk-python

Google ADK sdk-python

Pydantic AI sdk-python

Vercel AI SDK sdk-typescript