How to Use the LangSmith MCP in CrewAI

Q: Can a CrewAI agent monitor multiple LangSmith projects at once?

Yes. The 'Monitor' agent can first call langsmithlistprojects to get all project IDs. Then, it can loop through them, calling langsmithlistruns for each to check for errors or performance issues across your entire stack.

Equip your CrewAI agents with observability. Let one agent monitor LangSmith traces while another agent takes action on failures.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect LangSmith MCP to CrewAI

Create your Vinkius account to connect LangSmith to CrewAI 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 LangSmith with CrewAI

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Create a Dedicated 'Monitor' Agent

Assign an agent in your CrewAI setup the role of 'System Monitor'. Its only job is to use the `langsmith_list_runs` and `langsmith_list_projects` tools to watch for anomalies. It can continuously scan for high-latency traces or error flags across all your projects. This splits the workload perfectly. Your 'Worker' agents focus on their tasks, while the 'Monitor' agent provides a constant stream of performance data into the crew's shared memory. If it spots trouble, it can delegate a new task to another agent to handle it.

Build Autonomous Debugging Crews

When a run fails, your 'Monitor' agent can create a new task for a 'Debugger' agent. The task: investigate the failure using `langsmith_get_run` with the specific run ID. The 'Debugger' gets the full trace, including all inputs and outputs from the tool calls. Based on the trace data, the 'Debugger' can determine the root cause. Maybe it was a malformed input or a downstream API failure. It then reports its findings back to the crew, which can decide to retry the original task with corrected parameters or alert a human.

A LangSmith MCP Server for Your Whole Crew

Adding this LangSmith MCP server to your `crewai.Agent` is dead simple. Just pass the Vinkius URL in the `mcps` parameter. Now every agent in that crew—from the researcher to the writer—can access LangSmith tools. You can also use `tool_filter` for more control. Maybe only the 'Monitor' agent gets `langsmith_list_runs`, while the 'Debugger' is the only one with access to the more detailed `langsmith_get_run`. It lets you enforce roles at the tool level.

Setup guide

Set up LangSmith MCP in CrewAI

Prerequisites

Python 3.10+ installed
crewai package (pip install crewai)
Active Vinkius subscription with a valid endpoint token

1

Install CrewAI
Run pip install crewai to install the framework. MCP support is built-in via the mcps parameter.
2

Add the MCP URL to your agent
Pass your Vinkius endpoint directly to the mcps list. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. CrewAI handles tool discovery and caching automatically.
3

Kick off your crew
Create a Crew with your agent and tasks. Call crew.kickoff() — the agent will automatically invoke LangSmith tools as needed.

crew.py

from crewai import Agent, Task, Crew

agent = Agent(
    role="LangSmith Analyst",
    goal="Access and analyze LangSmith data via MCP.",
    backstory="Expert analyst with direct LangSmith access.",
    mcps=[
        "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
    ],
)

task = Task(
    description="List recent LangSmith transactions",
    agent=agent,
    expected_output="A summary of recent activity",
)

crew = Crew(agents=[agent], tasks=[task])
result = crew.kickoff()
print(result)

Get your connection token →

Prerequisites

Python 3.10+ installed
crewai + crewai-tools packages
Active Vinkius subscription with a valid endpoint token

1

Install dependencies
Run pip install crewai crewai-tools. The MCPServerAdapter handles lifecycle management and tool conversion.
2

Connect with MCPServerAdapter
Use MCPServerAdapter as a context manager with SseServerParameters pointing to your Vinkius endpoint. The adapter automatically manages connection lifecycle.
3

Assign tools and run
Pass the returned mcp_tools to your agent's tools parameter. The adapter converts MCP tools to native BaseTool objects compatible with all CrewAI agents.

crew.py

from crewai import Agent, Task, Crew
from crewai_tools import MCPServerAdapter
from mcp import SseServerParameters

server_params = SseServerParameters(
    url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
)

with MCPServerAdapter(server_params) as mcp_tools:
    agent = Agent(
        role="LangSmith Analyst",
        goal="Access and analyze LangSmith data via MCP.",
        backstory="Expert analyst with direct LangSmith access.",
        tools=mcp_tools,
    )

    task = Task(
        description="List recent LangSmith transactions",
        agent=agent,
        expected_output="A summary of recent activity",
    )

    crew = Crew(agents=[agent], tasks=[task])
    result = crew.kickoff()
    print(result)

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 LangSmith. 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 LangSmith 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 LangSmith MCP in CrewAI

When you define your agents, only provide the LangSmith MCP server URL to the agent that needs it. For more granular control, use the `MCPServerHTTP` class with a `tool_filter` to specify exactly which tools an agent can use.

Yes. The 'Monitor' agent can first call `langsmith_list_projects` to get all project IDs. Then, it can loop through them, calling `langsmith_list_runs` for each to check for errors or performance issues across your entire stack.

A common pattern is a two-agent crew. One agent, the 'Tracer', watches for new failed runs using `langsmith_list_runs`. When it finds one, it passes the run ID to a second agent, the 'Analyst', which uses `langsmith_get_run` to investigate and report the cause.

CrewAI is for building autonomous systems. A simple script can fetch data, but a crew can fetch data, analyze it, delegate a response, and verify the fix, all without human intervention. This MCP server gives your autonomous crew the senses it needs to do that.

The server processes only the metadata needed to interact with the LangSmith API, like run IDs and project names. Your LLM's inputs and outputs are not touched. Every request is handled in a separate, ephemeral Vinkius sandbox that is destroyed after execution, and all traffic is encrypted.

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