LambdaTest MCP server
LambdaTest's official remote MCP server: triage automation failures, pull command/network/console logs, run accessibility and visual checks from an agent.
The LambdaTest MCP Server is the official integration from LambdaTest (now TestMu AI) that connects an AI assistant to its cloud testing platform so an agent can triage failures and reason about test runs without leaving the editor. Its headline job is automation failure triaging: given a test ID, the agent fetches detailed test metadata, retrieves the Selenium command execution logs, inspects network traffic, and reads browser console output to surface the error-level messages and warnings that explain why a run broke, then proposes a fix in the same conversation.
Beyond automation logs the hosted server bundles several capabilities behind one endpoint: an Accessibility MCP that runs WCAG/ADA audits against public URLs or a locally served build, a SmartUI MCP for visual regression analysis across pixel, layout, DOM, and perceptual layers, and a HyperExecute MCP that analyzes a project, generates YAML config and runner commands, and answers documentation queries. It is a remote Streamable HTTP server reached at https://mcp.lambdatest.com/mcp and authenticated with OAuth 2.1 (you are redirected to sign in on first use), so there is no local binary to install — point any MCP client at the URL, or use the mcp-remote bridge for stdio-only clients.
Quick install
Copy-paste configs are provided for all 8 supported clients. Pick your client below.
Available tools
| Tool | Description |
|---|---|
| Automation Test Details | Fetches detailed information about a specific test ID from the LambdaTest cloud. |
| Automation Command Logs | Retrieves Selenium command execution logs for an automation run. |
| Automation Network Logs | Accesses test network logs for insight into browser traffic and behavior. |
| Automation Console Logs | Fetches browser console logs, surfacing error-level messages and warnings. |
| Accessibility Report | Generates a WCAG/ADA accessibility report for a public URL. |
| Accessibility Local Build | Builds and serves a local app to identify accessibility issues during development. |
| Accessibility Analyze App | Analyzes a tunneled or local app for accessibility violations. |
| SmartUI Visual Analysis | Runs visual regression analysis across pixel, layout, DOM, and perceptual layers. |
| HyperExecute Project Analysis | Analyzes a project and generates HyperExecute YAML configuration and runner commands. |
| HyperExecute Job Monitor | Monitors HyperExecute jobs and answers documentation queries for the runner. |
What you can do with it
Triage a failed cloud test from your IDE
Hand the agent a failing test ID and it pulls the command, network, and console logs from LambdaTest, reads the stack trace, and proposes a fix without you opening the dashboard.
Run an accessibility audit on a URL
Ask the agent to check a page for WCAG/ADA issues and it runs the Accessibility MCP against the public URL or a local build, returning a prioritized list of violations to fix.
FAQ
- Is it free?
- The MCP server itself is free to connect, but it operates against your LambdaTest (TestMu AI) account, so you need a plan that covers the automation, accessibility, SmartUI, or HyperExecute features the tools reach. Usage bills against your account as normal.
- Does it support remote/OAuth?
- Yes. It is a hosted Streamable HTTP server at https://mcp.lambdatest.com/mcp authenticated with OAuth 2.1 — you are redirected to sign in on first use. Stdio-only clients can connect through the mcp-remote bridge.
- What can the agent actually see?
- For automation it reads test metadata plus command, network, and console logs; the bundled Accessibility, SmartUI, and HyperExecute tools add WCAG audits, visual regression analysis, and config generation respectively.