How to Use the E2B MCP in LangChain

Q: Can LangChain manage multiple active E2B sandboxes at once?

Yes. Your LangChain agent can call createsandbox multiple times to run parallel tasks and use listsandboxes to track them. Each microVM runs in its own isolated environment.

Q: Can I use custom templates with this setup?

Yes. When your agent calls createsandbox, it can request templates like python3 or node. This gives your LangChain agent a pre-configured runtime environment instantly.

Run untrusted code safely in 150ms sandboxes directly from your LangChain chains and agents.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect E2B MCP to LangChain

Create your Vinkius account to connect E2B 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 E2B with LangChain

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Run untrusted code safely inside LangChain chains

Your LangChain agents often need to execute arbitrary code to solve math problems or analyze data. By integrating this MCP Server, your agent can spin up a secure Firecracker microVM using `create_sandbox` and run Python or JavaScript safely without risking your host machine. Once the execution finishes, the agent can fetch the output and feed it directly into the next link of the chain. You can trace this entire data flow in LangSmith, seeing exactly when the VM was created and when it was destroyed with `kill_sandbox`.

Chain sandbox lifecycles across complex agent runs

LangChain excels at multi-step reasoning where one tool's output feeds another. When your chain requires a persistent runtime, the agent can call `create_sandbox` to start a session, run multiple commands, and query `list_sandboxes` to keep track of active environments across execution steps. This setup lets your LangChain agents build on top of previous code executions in the same sandbox. When the agent completes its run, it calls `kill_sandbox` to tear down the microVM and avoid resource leaks.

Trace sandbox execution latency inside LangSmith

Debugging code execution in LangChain is notoriously tricky when you don't know why a step failed. Because this MCP Server integrates directly with your agent, every call to `create_sandbox` and command execution is captured as a distinct step in your LangSmith traces. You get clear visibility into the 150ms cold-start latency and can pinpoint exactly where a script failed. This takes the guesswork out of monitoring ephemeral environments across complex LangChain pipelines.

Setup guide

Set up E2B 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 E2B 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({
    "e2b-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 E2B 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 E2B. 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 E2B 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 E2B MCP in LangChain

You install the LangChain MCP adapter package and configure the server URL. Once connected, you pass the `create_sandbox` and `kill_sandbox` tools directly into your agent's tool list so it can execute code dynamically.

Yes. Your LangChain agent can call `create_sandbox` multiple times to run parallel tasks and use `list_sandboxes` to track them. Each microVM runs in its own isolated environment.

The default timeout is 300 seconds, but your LangChain agent can specify different parameters when calling `create_sandbox`. If an agent hangs, you can use `kill_sandbox` to force-terminate the microVM.

Yes. When your agent calls `create_sandbox`, it can request templates like python3 or node. This gives your LangChain agent a pre-configured runtime environment instantly.

Every script your LangChain agent runs is executed within an isolated Firecracker microVM. Once `kill_sandbox` is called, the entire filesystem is permanently wiped from the secure host, leaving no residual data behind.

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