Playwright Mcp

microsoft/playwright-mcpOfficial

Playwright MCP provides browser automation capabilities to LLMs through structured accessibility snapshots rather than screenshots, enabling deterministic web page interactions without requiring vision models. The server exposes Playwright's automation tools via the Model Context Protocol, making it suitable for exploratory automation, self-healing tests, and long-running autonomous workflows that benefit from persistent browser state and rich page structure introspection. It operates as a lightweight, token-efficient alternative to visual approaches, particularly valuable for specialized agentic loops requiring iterative reasoning over page content.

install copies · 90 days. Times the install command was copied on this site in the last 90 days, bots filtered.: 42
GitHub stars: 39.5K
tools: 23
transport: STDIO
last updated: Nov 9, 2026

View on GitHub

Install

From registry metadata

$ npx add-mcp @playwright/mcp --name playwright-mcp

Run in your terminal. add-mcp asks which agents to add it to (Claude Code, Codex, Cursor and more). Add -g to install it for every project.

Review the command, arguments and environment values before installing. MCP servers run with your local permissions.

Other ways

Claude Code

$ claude mcp add playwright-mcp -- npx -y @playwright/mcp

Run in your terminal. Replace YOUR_* placeholders with real values; add --scope user to install for every project.

Tools

Verified live · 23 tools

Verified live against the running server on Jun 10, 2026.

browser_closeClose the page

Close the page

No parameters. Call it with no arguments.

browser_resizeResize the browser window2 params

Resize the browser window

Parameters* required

height*number

Height of the browser window

width*number

Width of the browser window

browser_console_messagesReturns all console messages3 params

Returns all console messages

Parameters* required

allboolean

Return all console messages since the beginning of the session, not just since the last navigation. Defaults to false.

filenamestring

Filename to save the console messages to. If not provided, messages are returned as text.

level*string

Level of the console messages to return. Each level includes the messages of more severe levels. Defaults to "info".one of error · warning · info · debugdefault: info

browser_handle_dialogHandle a dialog2 params

Handle a dialog

Parameters* required

accept*boolean

Whether to accept the dialog.

promptTextstring

The text of the prompt in case of a prompt dialog.

browser_evaluateEvaluate JavaScript expression on page or element4 params

Evaluate JavaScript expression on page or element

Parameters* required

elementstring

Human-readable element description used to obtain permission to interact with the element

filenamestring

Filename to save the result to. If not provided, result is returned as text.

function*string

() => { /* code */ } or (element) => { /* code */ } when element is provided

targetstring

Exact target element reference from the page snapshot, or a unique element selector

browser_file_uploadUpload one or multiple files1 params

Upload one or multiple files

Parameters* required

pathsarray

The absolute paths to the files to upload. Can be single file or multiple files. If omitted, file chooser is cancelled.

browser_dropDrop files or MIME-typed data onto an element, as if dragged from outside the page. At least one of "paths" or "data" must be provided.4 params

Drop files or MIME-typed data onto an element, as if dragged from outside the page. At least one of "paths" or "data" must be provided.

Parameters* required

dataobject

Data to drop, as a map of MIME type to string value (e.g. {"text/plain": "hello", "text/uri-list": "https://example.com"}).

elementstring

Human-readable element description used to obtain permission to interact with the element

pathsarray

Absolute paths to files to drop onto the element.

target*string

Exact target element reference from the page snapshot, or a unique element selector

browser_fill_formFill multiple form fields1 params

Fill multiple form fields

Parameters* required

fields*array

Fields to fill in

browser_press_keyPress a key on the keyboard1 params

Press a key on the keyboard

Parameters* required

key*string

Name of the key to press or a character to generate, such as `ArrowLeft` or `a`

browser_typeType text into editable element5 params

Type text into editable element

Parameters* required

elementstring

Human-readable element description used to obtain permission to interact with the element

slowlyboolean

Whether to type one character at a time. Useful for triggering key handlers in the page. By default entire text is filled in at once.

submitboolean

Whether to submit entered text (press Enter after)

target*string

Exact target element reference from the page snapshot, or a unique element selector

text*string

Text to type into the element

browser_navigateNavigate to a URL1 params

Navigate to a URL

Parameters* required

url*string

The URL to navigate to

browser_navigate_backGo back to the previous page in the history

Go back to the previous page in the history

No parameters. Call it with no arguments.

browser_network_requestsReturns a numbered list of network requests since loading the page. Use browser_network_request with the number to get full details.3 params

Returns a numbered list of network requests since loading the page. Use browser_network_request with the number to get full details.

Parameters* required

filenamestring

Filename to save the network requests to. If not provided, requests are returned as text.

filterstring

Only return requests whose URL matches this regexp (e.g. "/api/.*user").

static*boolean

Whether to include successful static resources like images, fonts, scripts, etc. Defaults to false.default: false

browser_network_requestReturns full details (headers and body) of a single network request, or a single part if `part` is set. Use the number from browser_network_requests.3 params

Returns full details (headers and body) of a single network request, or a single part if `part` is set. Use the number from browser_network_requests.

Parameters* required

filenamestring

Filename to save the result to. If not provided, output is returned as text.

index*integer

1-based index of the request, as printed by browser_network_requests.

partstring

Return only this part of the request. Omit to return full details.one of request-headers · request-body · response-headers · response-body

browser_run_code_unsafeRun a Playwright code snippet. Unsafe: executes arbitrary JavaScript in the Playwright server process and is RCE-equivalent.2 params

Run a Playwright code snippet. Unsafe: executes arbitrary JavaScript in the Playwright server process and is RCE-equivalent.

Parameters* required

codestring

A JavaScript function containing Playwright code to execute. It will be invoked with a single argument, page, which you can use for any page interaction. For example: `async (page) => { await page....

filenamestring

Load code from the specified file. If both code and filename are provided, code will be ignored.

browser_take_screenshotTake a screenshot of the current page. You can't perform actions based on the screenshot, use browser_snapshot for actions.5 params

Take a screenshot of the current page. You can't perform actions based on the screenshot, use browser_snapshot for actions.

Parameters* required

elementstring

Human-readable element description used to obtain permission to interact with the element

filenamestring

File name to save the screenshot to. Defaults to `page-{timestamp}.{png|jpeg}` if not specified. Prefer relative file names to stay within the output directory.

fullPageboolean

When true, takes a screenshot of the full scrollable page, instead of the currently visible viewport. Cannot be used with element screenshots.

targetstring

Exact target element reference from the page snapshot, or a unique element selector

type*string

Image format for the screenshot. Default is png.one of png · jpegdefault: png

browser_snapshotCapture accessibility snapshot of the current page, this is better than screenshot4 params

Capture accessibility snapshot of the current page, this is better than screenshot

Parameters* required

boxesboolean

Include each element's bounding box as [box=x,y,width,height] in the snapshot. Coordinates are viewport-relative, in CSS pixels (Element.getBoundingClientRect)

depthnumber

Limit the depth of the snapshot tree

filenamestring

Save snapshot to markdown file instead of returning it in the response.

targetstring

Exact target element reference from the page snapshot, or a unique element selector

browser_clickPerform click on a web page5 params

Perform click on a web page

Parameters* required

buttonstring

Button to click, defaults to leftone of left · right · middle

doubleClickboolean

Whether to perform a double click instead of a single click

elementstring

Human-readable element description used to obtain permission to interact with the element

modifiersarray

Modifier keys to press

target*string

Exact target element reference from the page snapshot, or a unique element selector

browser_dragPerform drag and drop between two elements4 params

Perform drag and drop between two elements

Parameters* required

endElementstring

Human-readable target element description used to obtain the permission to interact with the element

endTarget*string

Exact target element reference from the page snapshot, or a unique element selector

startElementstring

Human-readable source element description used to obtain the permission to interact with the element

startTarget*string

Exact target element reference from the page snapshot, or a unique element selector

browser_hoverHover over element on page2 params

Hover over element on page

Parameters* required

elementstring

Human-readable element description used to obtain permission to interact with the element

target*string

Exact target element reference from the page snapshot, or a unique element selector

browser_select_optionSelect an option in a dropdown3 params

Select an option in a dropdown

Parameters* required

elementstring

Human-readable element description used to obtain permission to interact with the element

target*string

Exact target element reference from the page snapshot, or a unique element selector

values*array

Array of values to select in the dropdown. This can be a single value or multiple values.

browser_tabsList, create, close, or select a browser tab.3 params

List, create, close, or select a browser tab.

Parameters* required

action*string

Operation to performone of list · new · close · select

indexnumber

Tab index, used for close/select. If omitted for close, current tab is closed.

urlstring

URL to navigate to in the new tab, used for new.

browser_wait_forWait for text to appear or disappear or a specified time to pass3 params

Wait for text to appear or disappear or a specified time to pass

Parameters* required

textstring

The text to wait for

textGonestring

The text to wait for to disappear

timenumber

The time to wait in seconds

Readme

Playwright MCP

A Model Context Protocol (MCP) server that provides browser automation capabilities using Playwright. This server enables LLMs to interact with web pages through structured accessibility snapshots, bypassing the need for screenshots or visually-tuned models.

Playwright MCP vs Playwright CLI

This package provides MCP interface into Playwright. If you are using a coding agent, you might benefit from using the CLI+SKILLS instead.

CLI: Modern coding agents increasingly favor CLI–based workflows exposed as SKILLs over MCP because CLI invocations are more token-efficient: they avoid loading large tool schemas and verbose accessibility trees into the model context, allowing agents to act through concise, purpose-built commands. This makes CLI + SKILLs better suited for high-throughput coding agents that must balance browser automation with large codebases, tests, and reasoning within limited context windows.
Learn more about Playwright CLI with SKILLS.

View the full README on GitHub

Details

Repository: github.com/microsoft/playwright-mcp
Category: Web & Browser Automation
Package: @playwright/mcp
Transport: STDIO
Registry: MCP registry, active
Tools verified: Jun 10, 2026

Related Web & Browser Automation MCP servers

View all

Chrome DevToolsbenjaminr/chrome-devtools-mcpAn MCP Server for Chrome DevTools, following the Chrome DevTools Protocol. Integrates with Claude Desktop and Claude Code.297stars

Playwrightblackwhite084/playwright-plus-python-mcpExposes Playwright-based tools to automate browsing, capture content, and summarize stored notes via an MCP client.186stars