Official Microsoft Playwright MCP server that lets Claude drive a real browser through Playwright's accessibility tree for fast, deterministic web automation and testing.
by Microsoft · submitted by glorydavid03023·added 2026-06-02·
Launches and controls a real browser process that can navigate to arbitrary URLs and submit forms on your behalf., Treat any site the agent visits as untrusted; only allow automation against pages and actions you intend to run., Persisted browser profiles can keep cookies and logged-in sessions between runs, so isolate sensitive accounts.
Privacy notes
Page content, form input, and accessibility snapshots from visited sites are returned to the model context., Avoid driving the browser through pages that contain production credentials or private user data unless the profile is isolated.
Author
Microsoft
Submitted by
glorydavid03023
Claim status
unclaimed
Last verified
2026-06-02
Decision playbook
Review trust signals before you adopt
Signals are present but mixed. Use the checklist below to confirm the source and operational safety for your environment.
Compare context
Selected
0
Current score
78
Baseline
—
Delta
No baseline selected
No major trust-signal divergence detected in the current selection.
Source and provenance checks
Complete
Confirm ownership and provenance before trusting install instructions.
Source link availableRequired
Open the canonical repository and verify ownership.
Done
Source provenance statusRequired
Marked as source-backed.
Done
Metadata reviewed
Registry metadata indicates a reviewed listing.
Done
Safety and privacy checks
Complete
Validate risk disclosures before installation or API wiring.
Safety notes presentRequired
Review the listed safety guidance before running commands.
Done
Privacy notes presentRequired
Review data handling notes before connecting accounts or secrets.
Done
Trust level risk gateRequired
Trust level does not block evaluation.
Done
Package and install checks
Needs review
Check package metadata and artifact integrity signals.
Install payload available
Install or copy payload is available for review.
Done
Package verification flag
No package verification flag provided.
Pending
Checksum metadata
No checksum provided for downloaded artifact.
Pending
Compare-driven decision checks
Needs review
Use compare context to validate trade-offs before adoption.
Compare tray has multiple entries
Add at least one more entry to compare trust differences.
Current risk score 16/100. Use staged verification before broader rollout.
Risk 16
Pre-adoption checks
Validate source and review signals before any execution.
Confirm source provenanceRequired
Source URL/provenance metadata is present.
Done
Confirm metadata review state
Listing has review metadata.
Done
Verify install payload
Install/config payload exists and can be inspected.
Done
Security checks
Confirm safety, privacy, and package integrity signals.
Review safety notesRequired
Safety notes are present.
Done
Review privacy notesRequired
Privacy notes are present.
Done
Verify package integrity metadata
No package verification/checksum metadata.
Pending
Rollout
Adopt in controlled steps based on the selected plan.
Run in isolated sandbox firstRequired
Use a constrained sandbox and observe behavior across multiple tasks.
Pending
Roll out graduallyRequired
Roll out to a small cohort before wider usage.
Pending
Set monitoring and fallback
Define rollback path and monitor errors after adoption.
Pending
Evidence readiness
Evidence readiness matrix · balanced
Required evidence gates are covered (5/6 signals complete).
Risk 15
Source provenance
Present
Source repository/provenance is listed.
Required in this preset
Metadata review
Present
Review metadata is present.
Required in this preset
Safety notes
Present
Safety notes are present.
Required in this preset
Privacy notes
Present
Privacy notes are present.
Optional in this preset
Package integrity
Missing
Package integrity metadata is missing.
Optional in this preset
Install payload
Present
Install payload is available.
Required in this preset
Required evidence gates are covered for this preset.
Decision timeline
Decision timeline · balanced
5/6 steps complete with no blocking gaps for this preset.
Risk 14
triage
Confirm source provenanceRequired
Source/provenance metadata is available.
Done
triage
Check metadata review statusRequired
Review metadata is available.
Done
verify
Review safety notesRequired
Safety notes are available.
Done
verify
Review privacy notes
Privacy notes are available.
Done
verify
Validate package integrity metadata
Package integrity metadata is missing.
Pending
rollout
Verify install payload and commandsRequired
Install payload is available.
Done
No required blockers for this timeline preset.
Prerequisite readiness
Prerequisite readiness
3 prerequisites to line up before setup.
0/3 ready
Install & runtime2Network & hosting13 minutes
Safety & privacy surface
Safety & privacy surface
3 safety and 2 privacy notes across 3 risk areas. Review closely: credentials & tokens.
3 areas
SafetyExecution & processesLaunches and controls a real browser process that can navigate to arbitrary URLs and submit forms on your behalf.
SafetyExecution & processesTreat any site the agent visits as untrusted; only allow automation against pages and actions you intend to run.
SafetyCredentials & tokensPersisted browser profiles can keep cookies and logged-in sessions between runs, so isolate sensitive accounts.
PrivacyGeneralPage content, form input, and accessibility snapshots from visited sites are returned to the model context.
PrivacyCredentials & tokensAvoid driving the browser through pages that contain production credentials or private user data unless the profile is isolated.
Safety notes
Launches and controls a real browser process that can navigate to arbitrary URLs and submit forms on your behalf.
Treat any site the agent visits as untrusted; only allow automation against pages and actions you intend to run.
Persisted browser profiles can keep cookies and logged-in sessions between runs, so isolate sensitive accounts.
Privacy notes
Page content, form input, and accessibility snapshots from visited sites are returned to the model context.
Avoid driving the browser through pages that contain production credentials or private user data unless the profile is isolated.
Prerequisites
Node.js 18+ and npx available (verify with: npx --version)
Claude Code or Claude Desktop with MCP support
Internet access for the first run, which downloads the Playwright browser binaries
Playwright MCP connects Claude to a real browser through the official Microsoft Playwright engine. Instead of guessing at pixels, the server exposes the page's accessibility tree, so Claude can read structured page state and act on it with deterministic, element-level commands. That makes it well suited for end-to-end testing, form filling, scraping structured content, and reproducing user flows from natural language.
Features
Structured browser control built on the accessibility tree rather than screenshot or vision guessing.
Navigate, click, type, hover, select options, and submit forms with deterministic element targeting.
Capture accessibility snapshots of the current page for the model to reason over.
Works headless or headed, with support for the Chromium browsers Playwright manages.
Runs over stdio as a standard MCP server, so it drops into Claude Code and Claude Desktop with one command.
Maintained under the official Microsoft Playwright organization and tracks the upstream Playwright release line.
Use Cases
Drive end-to-end test scenarios described in plain language.
Fill and submit web forms during repetitive workflows.
Extract structured content from pages that require interaction before the data appears.
Reproduce and debug a reported user flow step by step.
Verify that a deployed change renders and behaves as expected in a real browser.
Installation
Claude Code
Make sure Node.js 18+ is installed (verify with npx --version).
Run: claude mcp add playwright -- npx @playwright/mcp@latest
Verify the server is registered: claude mcp list
Ask Claude to open a page to confirm the browser launches.
Claude Desktop
Open your Claude Desktop configuration file.
Add the Playwright server to the mcpServers section using the configuration below.
Restart Claude Desktop.
Confirm the server appears and ask Claude to navigate to a test URL.
Ask Claude to open a URL and describe what it sees from the accessibility snapshot.
"Open https://example.com and tell me the page title and the main heading."
Fill and submit a form
Drive a multi-field form using natural language; the server targets elements through the accessibility tree.
"Go to the contact form, enter a test name and email, type a short message, and submit it."
Reproduce a user flow for testing
Walk through a reported flow step by step to confirm behavior.
"Open the app, sign in with the test account, add an item to the cart, and confirm the cart count updates."
Security
The server controls a real browser that can reach any URL and perform actions like form submission; scope its use to pages and tasks you trust.
Treat visited pages as untrusted input. A malicious page could try to steer the agent, so review the actions it takes.
Persisted browser profiles can retain cookies and authenticated sessions between runs; isolate sensitive logins or use a clean profile.
Avoid running automation against pages that hold production credentials or private user data unless the environment is isolated.
Troubleshooting
Browser fails to launch on first run
The first launch may need to download a Playwright-managed Chromium build. Ensure the machine has internet access and enough disk space, then retry.
npx cannot find the package
Confirm Node.js 18+ and npx are installed (npx --version). Networks that block the npm registry will prevent npx @playwright/mcp@latest from resolving.
Server not listed in Claude Code
Re-run claude mcp add playwright -- npx @playwright/mcp@latest, then claude mcp list. Check that the command was added to the correct scope (project versus user) for the session you are running.
Actions target the wrong element
Ask Claude to take a fresh accessibility snapshot before acting. Page state can change after navigation or dynamic updates, so re-reading the tree improves targeting accuracy.
Show that Playwright MCP Server for Claude is listed on HeyClaude. Paste this Markdown into your README — it renders the badge and links back to this page.
[](https://heyclau.de/entry/mcp/playwright-mcp-server)
How it compares
Playwright MCP Server for Claude side by side with 2 alternatives on trust, install, platform support, and disclosed safety notes — all from reviewed registry metadata.
1 trust signal differ across this comparison (Submitter).
Official Microsoft Playwright MCP server that lets Claude drive a real browser through Playwright's accessibility tree for fast, deterministic web automation and testing.
Official Chrome DevTools MCP server that lets Claude inspect and debug a live Chrome browser, capturing performance traces, network activity, console messages, and DOM state through the Chrome DevTools Protocol.
✓Launches and controls a real browser process that can navigate to arbitrary URLs and submit forms on your behalf.
Treat any site the agent visits as untrusted; only allow automation against pages and actions you intend to run.
Persisted browser profiles can keep cookies and logged-in sessions between runs, so isolate sensitive accounts.
✓Browser pages can contain prompt-injection text in visible content, hidden HTML, ARIA labels, console output, network responses, or page metadata.
The server can navigate pages, click, hover, fill forms, drag, upload files, accept dialogs, dismiss dialogs, go through history, change viewport size, close tabs, and restart Firefox.
Enabling evaluate_script allows arbitrary JavaScript in page context; compromised instructions could read page data, modify the DOM, or interact with accessible browser APIs.
Enabling privileged context tools with the required Firefox system-access environment variable can expose privileged Firefox APIs and may extend beyond web-content sandbox boundaries.
Connect-existing mode can attach to a real browsing session with cookies, logins, active tabs, history, and saved state, so avoid it unless the profile is dedicated to automation.
Accepting insecure certificates weakens TLS validation and can hide man-in-the-middle or misconfiguration signals.
✓Launches and controls a real Chrome browser that can navigate to arbitrary URLs; only allow inspection of pages and actions you trust.
Treat inspected pages as untrusted input, since a hostile page could attempt to influence the agent through the content it surfaces.
Attaching to an existing Chrome profile can expose cookies and logged-in sessions, so prefer an isolated or dedicated profile for sensitive work.
Privacy notes
✓Page content, form input, and accessibility snapshots from visited sites are returned to the model context.
Avoid driving the browser through pages that contain production credentials or private user data unless the profile is isolated.
✓Page content, DOM snapshots, accessibility labels, console messages, network requests, headers, screenshots, uploaded file paths, extension names, Firefox prefs, and browser logs can be sent to the MCP client and model.
A regular Firefox profile can expose cookies, saved sessions, browsing history, account data, extensions, and private tabs to browser automation.
Screenshot save paths, uploaded files, downloaded artifacts, and model transcripts can reveal local project names, test data, credentials shown on pages, or customer information.
Use a separate profile, minimize enabled capabilities, avoid sensitive sites, and review provider/browser data handling before automating logged-in sessions.
✓Network capture and console output can include request payloads, tokens in URLs, and other sensitive runtime data returned to the model context.
Avoid debugging flows that carry production secrets or private user data unless the browser profile and environment are isolated.
Prerequisites
Node.js 18+ and npx available (verify with: npx --version)
Claude Code or Claude Desktop with MCP support
Internet access for the first run, which downloads the Playwright browser binaries
Node.js 20.19.0 or newer.
Local Firefox 100 or newer, or an explicit Firefox binary path passed through the documented CLI option.
A dedicated Firefox automation profile with no personal cookies, saved passwords, or regular browsing sessions.
Optional adb and Firefox for Android setup when using Android automation mode.
Node.js 18+ and npx available (verify with: npx --version)
A local installation of Google Chrome that the server can launch or attach to
Claude Code or Claude Desktop with MCP support
Install
claude mcp add playwright -- npx @playwright/mcp@latest
npx -y firefox-devtools-mcp@latest
claude mcp add chrome-devtools -- npx chrome-devtools-mcp@latest