How to Use the Percy MCP in LangChain

Q: How do I handle Percy build failures inside a LangChain runnable?

You run getbuilddetails within your chain to inspect the build state. If the state returns failed or contains unreviewed snapshots, your LangChain agent can catch the output and halt the deployment pipeline.

Q: Can a LangChain agent approve individual visual diffs using Percy?

Yes, the agent calls listsnapshots to fetch the visual diff percentages. If the differences fall within your acceptable threshold, the LangChain agent executes approvesnapshot to update the baseline.

Q: How does LangChain trace Percy MCP Server tool executions?

LangSmith logs every call to tools like listcomparisons and approvebuild. You get full visibility into the exact payload sent to the visual testing platform and the response latency.

Run visual regression checks inside your LangChain pipelines to catch UI breakage before deployment.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers — Included with Plan

Connect Percy MCP to LangChain

Create your Vinkius account to connect Percy to LangChain — we handle the hosting, security, and runtime updates so you don't have to. No server setup required.

GDPR Included with Plan

Setup Percy with LangChain

Ask AI about this MCP

ChatGPT

Claude

Perplexity Vinkius runs on Gemini

Gemini

Grok

Chain visual checks with LangChain agents

This Percy MCP Server exposes tools like `list_builds` and `get_build_details` directly to your LangChain runnable chains. Your LangChain agent inspects the state of active Percy visual test runs and decides whether to halt the pipeline or proceed based on actual pixel-diff counts. By passing these Percy tools into a LangChain ReAct agent, you let the model analyze the results of `list_comparisons` and execute `approve_snapshot` when visual changes match expected updates. LangSmith traces every step, giving you clear visibility into how your LangChain model handles Percy visual QA decisions.

Automate build approvals in LangGraph

The MCP Server lets your LangChain state machine manage complex deployment gates using `get_project_details` and `approve_build`. When a Percy visual regression run finishes, your LangChain agent checks the project configuration to verify if auto-approval is active. If the LangChain agent detects unreviewed Percy snapshots, it triggers `list_snapshots` to evaluate individual screen differences. The LangChain chain then makes a deterministic decision to approve the entire Percy build or flag it for manual developer review.

Multi-server visual routing in LangChain

Using the LangChain `MultiServerMCPClient`, you can combine this Percy visual testing tool with your other development APIs. Your LangChain agent queries `list_projects` to locate the correct Percy target repository and matches it against incoming pull request metadata. The LangChain agent calls `list_browsers` to confirm Percy target coverage before analyzing visual diff percentages. It acts as an autonomous LangChain QA gatekeeper, coordinating Percy visual data with your database and Slack channels in a single execution thread.

Setup guide

Set up Percy 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 Percy 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({
    "percy-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 Percy 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 Percy. 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 Percy 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 Percy MCP in LangChain

You run `get_build_details` within your chain to inspect the build state. If the state returns failed or contains unreviewed snapshots, your LangChain agent can catch the output and halt the deployment pipeline.

Yes, the agent calls `list_snapshots` to fetch the visual diff percentages. If the differences fall within your acceptable threshold, the LangChain agent executes `approve_snapshot` to update the baseline.

LangSmith logs every call to tools like `list_comparisons` and `approve_build`. You get full visibility into the exact payload sent to the visual testing platform and the response latency.

You instantiate the client with the server's HTTP transport URL and call `get_tools()`. This registers the visual regression tools alongside your other active servers in the LangChain agent.

No, this server only handles visual snapshots, build metadata, and project configurations. Your actual application source code remains entirely within your local environment or CI runner.

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