Playwright MCP vs Usable Browser Agent
A side-by-side look at the Playwright MCP and Usable Browser Agent MCP servers: transport, auth, tools, quality signals, and the config to install either one in Claude, Cursor, Windsurf or VS Code.
Playwright MCPBrowser Automation · Local subprocess (stdio)Official Microsoft Playwright MCP server, enabling LLMs to interact with web pages through structured accessibility snapshots89ExcellentAuth: no auth requiredVerdict: choose Playwright MCP for browser automation work as a local process; choose Usable Browser Agent for browser automation. You can also run both side by side in one mcpServers block.
Which MCP server should you choose?
Choose Playwright MCP when
- You need dedicated capabilities in the Browser Automation domain.
- You prefer a local stdio subprocess transport.
- Its auth model fits you: No auth required.
- You want tools like
browser_close, browser_resize, browser_console_messages.
Choose Usable Browser Agent when
- You need dedicated capabilities in the Browser Automation domain.
- You have the keys it needs:
UBA_PORT, UBA_MEMORY_FILE, UBA_SECRETS_FILE. - You want tools like
browser_workflow_recall, browser_workflow_remember, browser_workflow_forget.
Feature & specification comparison
Rows where the two servers differ are marked.
| Spec | Playwright MCP | |
|---|---|---|
| Category | Browser Automation | Browser Automation |
| Quality signal | 89/100 · Excellent | 45/100 · Fair |
| Install command | Confirmed | Not confirmed yet |
| Transport | Local subprocess (stdio) | — |
| Auth | No auth required | Not declared |
| Pricing | Free | Not declared |
| Required env vars | None listed | UBA_PORTUBA_MEMORY_FILEUBA_SECRETS_FILE |
| Verified owner | Verified | Community listing |
| GitHub stars | 38,407 | 0 |
| Directory activity | 6 views 0 installs 0 upvotes | 1 views 0 installs 0 upvotes |
| Tools | 25 | 34 |
| Full listing | View Playwright MCP | View Usable Browser Agent |
Tools & capabilities
Playwright MCP tools 25
browser_closeClose the pagebrowser_resizeResize the browser windowbrowser_console_messagesReturns all console messagesbrowser_handle_dialogHandle a dialogbrowser_emulate_mediaEmulate CSS media features for the page, for example switch between the light and dark color scheme. Omitted parameters are left unchanged; null clears an override.browser_evaluateEvaluate JavaScript expression on page or elementbrowser_file_uploadUpload one or multiple filesbrowser_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.browser_findSearch the accessibility snapshot of the current page for text or a regular expression. Returns matching snapshot nodes with a few lines of surrounding context (like search snippets), each shown under its path from the root of the tree, which is cheaper than capturing the whole snapshot when you only need to locate an element and its ref.browser_fill_formFill multiple form fieldsbrowser_press_keyPress a key on the keyboardbrowser_typeType text into editable element- +13 more on the listing page
Usable Browser Agent tools 34
browser_workflow_recallSearch local workflow memory before starting a repeated/changing-site taskbrowser_workflow_rememberSave verified reusable workflow steps, pitfalls, and durable UI cuesbrowser_workflow_forgetTombstone stale, sensitive, or incorrect workflow memorybrowser_snapshotAccessibility/DOM tree with `[ref=eN]` handles — the primary perception toolbrowser_navigateNavigationbrowser_clickClick by ref — supports right/middle button, double-click, and ctrl/shift/alt/meta modifiersbrowser_dragDrag one element onto another (kanban, sortable lists, sliders)browser_typeType into a field by ref (optional `submit` = press Enter)browser_fill_formFill many fields `[{ref,text}]` in one callbrowser_fill_secretFill a username/password from a server-side secret aliasbrowser_loginAtomically fill username/password from a server-side secret alias and submitbrowser_selectChoose a `<select>` option by label/value- +22 more on the listing page
Ready-to-paste client configs
Paste either block (or both) into your client config, e.g. claude_desktop_config.json or ~/.cursor/mcp.json. We only show a config when the install command is confirmed.
Playwright MCP
{
"mcpServers": {
"microsoft-playwright-mcp": {
"command": "npx",
"args": [
"-y",
"@playwright/mcp@latest"
]
}
}
}
Usable Browser Agent
No confirmed install command for Usable Browser Agent yet — follow the setup steps linked from its listing page rather than a guessed package name.
Frequently asked questions
Playwright MCP is categorized under Browser Automation and uses a local stdio subprocess. Usable Browser Agent belongs to Browser Automation and uses an install path we have not confirmed yet. Pick Playwright MCP for browser automation work and Usable Browser Agent for browser automation.
Copy the mcpServers block from the config section on this page into your client's config (for example claude_desktop_config.json or ~/.cursor/mcp.json), then fully restart the client. Usable Browser Agent has no confirmed install command yet — follow its own setup instructions.
Playwright MCP is listed as free. Auth: no auth required. Usable Browser Agent has no declared pricing — check the project's own site. If a server needs environment variables such as API keys, define them under the env key in your client's MCP config.
Yes. MCP clients support multiple servers: add microsoft-playwright-mcp and usable-browser-agent as separate keys in the same mcpServers object, and your assistant will call tools from either one.
On AllMCPs, Playwright MCP has a quality signal of 89/100 (Excellent) with 6 views, 0 installs and 38,407 GitHub stars. Usable Browser Agent has 45/100 (Fair) with 1 views, 0 installs and 0 GitHub stars.