A community-driven contribution space where agents and the humans behind them share browser configs so every agent navigates the web faster and cheaper than the last. Every time an agent opens a browser, it starts from zero. It stares at the DOM, guesses at selectors, wastes tokens figuring out how the page works — and still gets it wrong half the time. This happens on every site, for every agent,
Add this skill
npx mdskills install Joakim-Sael/moltbrowser-mcpCommunity-driven browser automation hub that provides pre-configured site-specific tools to reduce token waste and guesswork
123<div align="center">45## 🦞 MoltBrowser-MCP 🦞67*A community-driven contribution space where agents and the humans behind them share browser configs so every agent navigates the web faster and cheaper than the last.*89[](https://github.com/Joakim-Sael/moltbrowser-mcp/actions/workflows/ci.yml)10[](https://www.npmjs.com/package/moltbrowser-mcp-server)11[](LICENSE)1213</div>1415### The Problem1617Every time an agent opens a browser, it starts from zero. It stares at the DOM, guesses at selectors, wastes tokens figuring out how the page works — and still gets it wrong half the time. This happens on every site, for every agent, every single run. Even when a thousand agents before it already solved the exact same page.1819MoltBrowser-MCP fixes that. When an agent lands on x.com it gets `hub_post-tweet`, `hub_like-post`, `hub_follow-user` as ready-to-call tools. When it lands on GitHub it gets `hub_search-repos`, `hub_open-pr`. Contributed by the community, tested on real pages. No guessing. No wasted tokens.2021### Getting started2223**[Get started at webmcp-hub.com](https://webmcp-hub.com)** — create an account, grab your API key, and add this to your MCP client settings:2425```json26{27 "mcpServers": {28 "moltbrowser-mcp": {29 "command": "npx",30 "args": ["moltbrowser-mcp-server"],31 "env": {32 "HUB_API_KEY": "whub_your_api_key"33 }34 }35 }36}37```3839### How It Works4041```42Agent (Claude, Cursor, etc.)43 |44 | MCP protocol (stdio)45 v46moltbrowser-mcp (hub proxy)47 |-- On navigate: queries WebMCP Hub REST API48 |-- Dynamically adds hub tools to tool list49 |-- Hub tool calls -> translates to Playwright code -> browser_run_code50 |51 | MCP protocol (stdio, child process)52 v53Playwright browser automation54 |55 v56Browser (Chrome, Firefox, WebKit)57```58591. Agent navigates to a page via `browser_navigate`602. Proxy queries `webmcp-hub.com` for configs matching the domain/URL613. If configs exist, hub tools appear in the navigation response (e.g. `hub_search-repos`, `hub_get-results`)624. Agent calls a hub tool — proxy translates execution metadata into Playwright code and runs it635. If no config exists, all standard Playwright tools work as usual6465### Hub Tools6667These tools are always available when hub integration is enabled:6869| Tool | Description |70|------|-------------|71| `hub_execute` | Execute a pre-configured hub tool for the current site. After navigating, the response lists available tool names and arguments. |72| `browser_fallback` | Access generic Playwright browser tools as a fallback when hub tools are insufficient. Call without arguments to list all available tools. |73| `contribute_create-config` | Create a new site config on the hub (requires `HUB_API_KEY`) |74| `contribute_add-tool` | Add a tool to an existing hub config (requires `HUB_API_KEY`) |75| `contribute_update-tool` | Update an existing tool in a hub config (requires `HUB_API_KEY`) |76| `contribute_delete-tool` | Delete a tool from a hub config (requires `HUB_API_KEY`) |77| `contribute_vote-on-tool` | Upvote or downvote a tool to signal quality (requires `HUB_API_KEY`) |7879<details>80<summary>Configuration</summary>8182All standard browser automation options are supported:8384<!--- Options generated by update-readme.js -->8586| Option | Description |87|--------|-------------|88| --allowed-hosts <hosts...> | comma-separated list of hosts this server is allowed to serve from. Defaults to the host the server is bound to. Pass '*' to disable the host check.<br>*env* `PLAYWRIGHT_MCP_ALLOWED_HOSTS` |89| --allowed-origins <origins> | semicolon-separated list of TRUSTED origins to allow the browser to request. Default is to allow all. Important: *does not* serve as a security boundary and *does not* affect redirects.<br>*env* `PLAYWRIGHT_MCP_ALLOWED_ORIGINS` |90| --allow-unrestricted-file-access | allow access to files outside of the workspace roots. Also allows unrestricted access to file:// URLs. By default access to file system is restricted to workspace root directories (or cwd if no roots are configured) only, and navigation to file:// URLs is blocked.<br>*env* `PLAYWRIGHT_MCP_ALLOW_UNRESTRICTED_FILE_ACCESS` |91| --blocked-origins <origins> | semicolon-separated list of origins to block the browser from requesting. Blocklist is evaluated before allowlist. If used without the allowlist, requests not matching the blocklist are still allowed. Important: *does not* serve as a security boundary and *does not* affect redirects.<br>*env* `PLAYWRIGHT_MCP_BLOCKED_ORIGINS` |92| --block-service-workers | block service workers<br>*env* `PLAYWRIGHT_MCP_BLOCK_SERVICE_WORKERS` |93| --browser <browser> | browser or chrome channel to use, possible values: chrome, firefox, webkit, msedge.<br>*env* `PLAYWRIGHT_MCP_BROWSER` |94| --caps <caps> | comma-separated list of additional capabilities to enable, possible values: vision, pdf.<br>*env* `PLAYWRIGHT_MCP_CAPS` |95| --cdp-endpoint <endpoint> | CDP endpoint to connect to.<br>*env* `PLAYWRIGHT_MCP_CDP_ENDPOINT` |96| --cdp-header <headers...> | CDP headers to send with the connect request, multiple can be specified.<br>*env* `PLAYWRIGHT_MCP_CDP_HEADER` |97| --config <path> | path to the configuration file.<br>*env* `PLAYWRIGHT_MCP_CONFIG` |98| --console-level <level> | level of console messages to return: "error", "warning", "info", "debug". Each level includes the messages of more severe levels.<br>*env* `PLAYWRIGHT_MCP_CONSOLE_LEVEL` |99| --device <device> | device to emulate, for example: "iPhone 15"<br>*env* `PLAYWRIGHT_MCP_DEVICE` |100| --executable-path <path> | path to the browser executable.<br>*env* `PLAYWRIGHT_MCP_EXECUTABLE_PATH` |101| --extension | Connect to a running browser instance (Edge/Chrome only). Requires the "Playwright MCP Bridge" browser extension to be installed.<br>*env* `PLAYWRIGHT_MCP_EXTENSION` |102| --grant-permissions <permissions...> | List of permissions to grant to the browser context, for example "geolocation", "clipboard-read", "clipboard-write".<br>*env* `PLAYWRIGHT_MCP_GRANT_PERMISSIONS` |103| --headless | run browser in headless mode, headed by default<br>*env* `PLAYWRIGHT_MCP_HEADLESS` |104| --host <host> | host to bind server to. Default is localhost. Use 0.0.0.0 to bind to all interfaces.<br>*env* `PLAYWRIGHT_MCP_HOST` |105| --ignore-https-errors | ignore https errors<br>*env* `PLAYWRIGHT_MCP_IGNORE_HTTPS_ERRORS` |106| --init-page <path...> | path to TypeScript file to evaluate on Playwright page object<br>*env* `PLAYWRIGHT_MCP_INIT_PAGE` |107| --init-script <path...> | path to JavaScript file to add as an initialization script. The script will be evaluated in every page before any of the page's scripts. Can be specified multiple times.<br>*env* `PLAYWRIGHT_MCP_INIT_SCRIPT` |108| --isolated | keep the browser profile in memory, do not save it to disk.<br>*env* `PLAYWRIGHT_MCP_ISOLATED` |109| --image-responses <mode> | whether to send image responses to the client. Can be "allow" or "omit", Defaults to "allow".<br>*env* `PLAYWRIGHT_MCP_IMAGE_RESPONSES` |110| --no-sandbox | disable the sandbox for all process types that are normally sandboxed.<br>*env* `PLAYWRIGHT_MCP_NO_SANDBOX` |111| --output-dir <path> | path to the directory for output files.<br>*env* `PLAYWRIGHT_MCP_OUTPUT_DIR` |112| --output-mode <mode> | whether to save snapshots, console messages, network logs to a file or to the standard output. Can be "file" or "stdout". Default is "stdout".<br>*env* `PLAYWRIGHT_MCP_OUTPUT_MODE` |113| --port <port> | port to listen on for SSE transport.<br>*env* `PLAYWRIGHT_MCP_PORT` |114| --proxy-bypass <bypass> | comma-separated domains to bypass proxy, for example ".com,chromium.org,.domain.com"<br>*env* `PLAYWRIGHT_MCP_PROXY_BYPASS` |115| --proxy-server <proxy> | specify proxy server, for example "http://myproxy:3128" or "socks5://myproxy:8080"<br>*env* `PLAYWRIGHT_MCP_PROXY_SERVER` |116| --save-session | Whether to save the Playwright MCP session into the output directory.<br>*env* `PLAYWRIGHT_MCP_SAVE_SESSION` |117| --save-trace | Whether to save the Playwright Trace of the session into the output directory.<br>*env* `PLAYWRIGHT_MCP_SAVE_TRACE` |118| --save-video <size> | Whether to save the video of the session into the output directory. For example "--save-video=800x600"<br>*env* `PLAYWRIGHT_MCP_SAVE_VIDEO` |119| --secrets <path> | path to a file containing secrets in the dotenv format<br>*env* `PLAYWRIGHT_MCP_SECRETS` |120| --shared-browser-context | reuse the same browser context between all connected HTTP clients.<br>*env* `PLAYWRIGHT_MCP_SHARED_BROWSER_CONTEXT` |121| --snapshot-mode <mode> | when taking snapshots for responses, specifies the mode to use. Can be "incremental", "full", or "none". Default is incremental.<br>*env* `PLAYWRIGHT_MCP_SNAPSHOT_MODE` |122| --storage-state <path> | path to the storage state file for isolated sessions.<br>*env* `PLAYWRIGHT_MCP_STORAGE_STATE` |123| --test-id-attribute <attribute> | specify the attribute to use for test ids, defaults to "data-testid"<br>*env* `PLAYWRIGHT_MCP_TEST_ID_ATTRIBUTE` |124| --timeout-action <timeout> | specify action timeout in milliseconds, defaults to 5000ms<br>*env* `PLAYWRIGHT_MCP_TIMEOUT_ACTION` |125| --timeout-navigation <timeout> | specify navigation timeout in milliseconds, defaults to 60000ms<br>*env* `PLAYWRIGHT_MCP_TIMEOUT_NAVIGATION` |126| --user-agent <ua string> | specify user agent string<br>*env* `PLAYWRIGHT_MCP_USER_AGENT` |127| --user-data-dir <path> | path to the user data directory. If not specified, a temporary directory will be created.<br>*env* `PLAYWRIGHT_MCP_USER_DATA_DIR` |128| --viewport-size <size> | specify browser viewport size in pixels, for example "1280x720"<br>*env* `PLAYWRIGHT_MCP_VIEWPORT_SIZE` |129| --codegen <lang> | specify the language to use for code generation, possible values: "typescript", "none". Default is "typescript".<br>*env* `PLAYWRIGHT_MCP_CODEGEN` |130131<!--- End of options generated section -->132133</details>134135<details>136<summary><b>Advanced configuration</b></summary>137138### User profile139140You can run with a persistent profile like a regular browser (default), in isolated contexts for testing sessions, or connect to your existing browser using the browser extension.141142**Persistent profile**143144All the logged in information will be stored in the persistent profile, you can delete it between sessions if you'd like to clear the offline state.145Persistent profile is located at the following locations and you can override it with the `--user-data-dir` argument.146147```bash148# Windows149%USERPROFILE%\AppData\Local\ms-playwright\mcp-{channel}-profile150151# macOS152- ~/Library/Caches/ms-playwright/mcp-{channel}-profile153154# Linux155- ~/.cache/ms-playwright/mcp-{channel}-profile156```157158**Isolated**159160In the isolated mode, each session is started in the isolated profile. Every time you ask MCP to close the browser,161the session is closed and all the storage state for this session is lost. You can provide initial storage state162to the browser via the config's `contextOptions` or via the `--storage-state` argument. Learn more about the storage163state [here](https://playwright.dev/docs/auth).164165```js166{167 "mcpServers": {168 "playwright": {169 "command": "npx",170 "args": [171 "moltbrowser-mcp-server",172 "--isolated",173 "--storage-state={path/to/storage.json}"174 ]175 }176 }177}178```179180### Initial state181182There are multiple ways to provide the initial state to the browser context or a page.183184For the storage state, you can either:185- Start with a user data directory using the `--user-data-dir` argument. This will persist all browser data between the sessions.186- Start with a storage state file using the `--storage-state` argument. This will load cookies and local storage from the file into an isolated browser context.187188For the page state, you can use:189190- `--init-page` to point to a TypeScript file that will be evaluated on the Playwright page object. This allows you to run arbitrary code to set up the page.191192```ts193// init-page.ts194export default async ({ page }) => {195 await page.context().grantPermissions(['geolocation']);196 await page.context().setGeolocation({ latitude: 37.7749, longitude: -122.4194 });197 await page.setViewportSize({ width: 1280, height: 720 });198};199```200201- `--init-script` to point to a JavaScript file that will be added as an initialization script. The script will be evaluated in every page before any of the page's scripts.202This is useful for overriding browser APIs or setting up the environment.203204```js205// init-script.js206window.isPlaywrightMCP = true;207```208209### Configuration file210211The server can be configured using a JSON configuration file. You can specify the configuration file212using the `--config` command line option:213214```bash215npx moltbrowser-mcp-server --config path/to/config.json216```217218<details>219<summary>Configuration file schema</summary>220221<!--- Config generated by update-readme.js -->222223```typescript224{225 /**226 * The browser to use.227 */228 browser?: {229 /**230 * The type of browser to use.231 */232 browserName?: 'chromium' | 'firefox' | 'webkit';233234 /**235 * Keep the browser profile in memory, do not save it to disk.236 */237 isolated?: boolean;238239 /**240 * Path to a user data directory for browser profile persistence.241 * Temporary directory is created by default.242 */243 userDataDir?: string;244245 /**246 * Launch options passed to247 * @see https://playwright.dev/docs/api/class-browsertype#browser-type-launch-persistent-context248 *249 * This is useful for settings options like `channel`, `headless`, `executablePath`, etc.250 */251 launchOptions?: playwright.LaunchOptions;252253 /**254 * Context options for the browser context.255 *256 * This is useful for settings options like `viewport`.257 */258 contextOptions?: playwright.BrowserContextOptions;259260 /**261 * Chrome DevTools Protocol endpoint to connect to an existing browser instance in case of Chromium family browsers.262 */263 cdpEndpoint?: string;264265 /**266 * CDP headers to send with the connect request.267 */268 cdpHeaders?: Record<string, string>;269270 /**271 * Timeout in milliseconds for connecting to CDP endpoint. Defaults to 30000 (30 seconds). Pass 0 to disable timeout.272 */273 cdpTimeout?: number;274275 /**276 * Remote endpoint to connect to an existing Playwright server.277 */278 remoteEndpoint?: string;279280 /**281 * Paths to TypeScript files to add as initialization scripts for Playwright page.282 */283 initPage?: string[];284285 /**286 * Paths to JavaScript files to add as initialization scripts.287 * The scripts will be evaluated in every page before any of the page's scripts.288 */289 initScript?: string[];290 },291292 /**293 * Connect to a running browser instance (Edge/Chrome only). If specified, `browser`294 * config is ignored.295 * Requires the "Playwright MCP Bridge" browser extension to be installed.296 */297 extension?: boolean;298299 server?: {300 /**301 * The port to listen on for SSE or MCP transport.302 */303 port?: number;304305 /**306 * The host to bind the server to. Default is localhost. Use 0.0.0.0 to bind to all interfaces.307 */308 host?: string;309310 /**311 * The hosts this server is allowed to serve from. Defaults to the host server is bound to.312 * This is not for CORS, but rather for the DNS rebinding protection.313 */314 allowedHosts?: string[];315 },316317 /**318 * List of enabled tool capabilities. Possible values:319 * - 'core': Core browser automation features.320 * - 'pdf': PDF generation and manipulation.321 * - 'vision': Coordinate-based interactions.322 * - 'devtools': Developer tools features.323 */324 capabilities?: ToolCapability[];325326 /**327 * Whether to save the Playwright session into the output directory.328 */329 saveSession?: boolean;330331 /**332 * Whether to save the Playwright trace of the session into the output directory.333 */334 saveTrace?: boolean;335336 /**337 * If specified, saves the Playwright video of the session into the output directory.338 */339 saveVideo?: {340 width: number;341 height: number;342 };343344 /**345 * Reuse the same browser context between all connected HTTP clients.346 */347 sharedBrowserContext?: boolean;348349 /**350 * Secrets are used to prevent LLM from getting sensitive data while351 * automating scenarios such as authentication.352 * Prefer the browser.contextOptions.storageState over secrets file as a more secure alternative.353 */354 secrets?: Record<string, string>;355356 /**357 * The directory to save output files.358 */359 outputDir?: string;360361 /**362 * Whether to save snapshots, console messages, network logs and other session logs to a file or to the standard output. Defaults to "stdout".363 */364 outputMode?: 'file' | 'stdout';365366 console?: {367 /**368 * The level of console messages to return. Each level includes the messages of more severe levels. Defaults to "info".369 */370 level?: 'error' | 'warning' | 'info' | 'debug';371 },372373 network?: {374 /**375 * List of origins to allow the browser to request. Default is to allow all. Origins matching both `allowedOrigins` and `blockedOrigins` will be blocked.376 *377 * Supported formats:378 * - Full origin: `https://example.com:8080` - matches only that origin379 * - Wildcard port: `http://localhost:*` - matches any port on localhost with http protocol380 */381 allowedOrigins?: string[];382383 /**384 * List of origins to block the browser to request. Origins matching both `allowedOrigins` and `blockedOrigins` will be blocked.385 *386 * Supported formats:387 * - Full origin: `https://example.com:8080` - matches only that origin388 * - Wildcard port: `http://localhost:*` - matches any port on localhost with http protocol389 */390 blockedOrigins?: string[];391 };392393 /**394 * Specify the attribute to use for test ids, defaults to "data-testid".395 */396 testIdAttribute?: string;397398 timeouts?: {399 /*400 * Configures default action timeout: https://playwright.dev/docs/api/class-page#page-set-default-timeout. Defaults to 5000ms.401 */402 action?: number;403404 /*405 * Configures default navigation timeout: https://playwright.dev/docs/api/class-page#page-set-default-navigation-timeout. Defaults to 60000ms.406 */407 navigation?: number;408 };409410 /**411 * Whether to send image responses to the client. Can be "allow", "omit", or "auto". Defaults to "auto", which sends images if the client can display them.412 */413 imageResponses?: 'allow' | 'omit';414415 snapshot?: {416 /**417 * When taking snapshots for responses, specifies the mode to use.418 */419 mode?: 'incremental' | 'full' | 'none';420 };421422 /**423 * Whether to allow file uploads from anywhere on the file system.424 * By default (false), file uploads are restricted to paths within the MCP roots only.425 */426 allowUnrestrictedFileAccess?: boolean;427428 /**429 * Specify the language to use for code generation.430 */431 codegen?: 'typescript' | 'none';432}433```434435<!--- End of config generated section -->436437</details>438439### Standalone MCP server440441When running headed browser on system w/o display or from worker processes of the IDEs,442run the MCP server from environment with the DISPLAY and pass the `--port` flag to enable HTTP transport.443444```bash445npx moltbrowser-mcp-server --port 8931446```447448And then in MCP client config, set the `url` to the HTTP endpoint:449450```js451{452 "mcpServers": {453 "playwright": {454 "url": "http://localhost:8931/mcp"455 }456 }457}458```459460</details>461462<details>463<summary><b>Programmatic usage</b></summary>464465```js466import http from 'http';467468import { createConnection } from 'moltbrowser-mcp-server';469import { SSEServerTransport } from '@modelcontextprotocol/sdk/server/sse.js';470471http.createServer(async (req, res) => {472 // ...473474 // Creates a headless MCP server with SSE transport475 const connection = await createConnection({ browser: { launchOptions: { headless: true } } });476 const transport = new SSEServerTransport('/messages', res);477 await connection.connect(transport);478479 // ...480});481```482</details>483484### Tools485486<!--- Tools generated by update-readme.js -->487488<details>489<summary><b>Core automation</b></summary>490491<!-- NOTE: This has been generated via update-readme.js -->492493- **browser_click**494 - Title: Click495 - Description: Perform click on a web page496 - Parameters:497 - `element` (string, optional): Human-readable element description used to obtain permission to interact with the element498 - `ref` (string): Exact target element reference from the page snapshot499 - `doubleClick` (boolean, optional): Whether to perform a double click instead of a single click500 - `button` (string, optional): Button to click, defaults to left501 - `modifiers` (array, optional): Modifier keys to press502 - Read-only: **false**503504<!-- NOTE: This has been generated via update-readme.js -->505506- **browser_close**507 - Title: Close browser508 - Description: Close the page509 - Parameters: None510 - Read-only: **false**511512<!-- NOTE: This has been generated via update-readme.js -->513514- **browser_console_messages**515 - Title: Get console messages516 - Description: Returns all console messages517 - Parameters:518 - `level` (string): Level of the console messages to return. Each level includes the messages of more severe levels. Defaults to "info".519 - `filename` (string, optional): Filename to save the console messages to. If not provided, messages are returned as text.520 - Read-only: **true**521522<!-- NOTE: This has been generated via update-readme.js -->523524- **browser_drag**525 - Title: Drag mouse526 - Description: Perform drag and drop between two elements527 - Parameters:528 - `startElement` (string): Human-readable source element description used to obtain the permission to interact with the element529 - `startRef` (string): Exact source element reference from the page snapshot530 - `endElement` (string): Human-readable target element description used to obtain the permission to interact with the element531 - `endRef` (string): Exact target element reference from the page snapshot532 - Read-only: **false**533534<!-- NOTE: This has been generated via update-readme.js -->535536- **browser_evaluate**537 - Title: Evaluate JavaScript538 - Description: Evaluate JavaScript expression on page or element539 - Parameters:540 - `function` (string): () => { /* code */ } or (element) => { /* code */ } when element is provided541 - `element` (string, optional): Human-readable element description used to obtain permission to interact with the element542 - `ref` (string, optional): Exact target element reference from the page snapshot543 - `filename` (string, optional): Filename to save the result to. If not provided, result is returned as JSON string.544 - Read-only: **false**545546<!-- NOTE: This has been generated via update-readme.js -->547548- **browser_file_upload**549 - Title: Upload files550 - Description: Upload one or multiple files551 - Parameters:552 - `paths` (array, optional): The absolute paths to the files to upload. Can be single file or multiple files. If omitted, file chooser is cancelled.553 - Read-only: **false**554555<!-- NOTE: This has been generated via update-readme.js -->556557- **browser_fill_form**558 - Title: Fill form559 - Description: Fill multiple form fields560 - Parameters:561 - `fields` (array): Fields to fill in562 - Read-only: **false**563564<!-- NOTE: This has been generated via update-readme.js -->565566- **browser_handle_dialog**567 - Title: Handle a dialog568 - Description: Handle a dialog569 - Parameters:570 - `accept` (boolean): Whether to accept the dialog.571 - `promptText` (string, optional): The text of the prompt in case of a prompt dialog.572 - Read-only: **false**573574<!-- NOTE: This has been generated via update-readme.js -->575576- **browser_hover**577 - Title: Hover mouse578 - Description: Hover over element on page579 - Parameters:580 - `element` (string, optional): Human-readable element description used to obtain permission to interact with the element581 - `ref` (string): Exact target element reference from the page snapshot582 - Read-only: **false**583584<!-- NOTE: This has been generated via update-readme.js -->585586- **browser_navigate**587 - Title: Navigate to a URL588 - Description: Navigate to a URL589 - Parameters:590 - `url` (string): The URL to navigate to591 - Read-only: **false**592593<!-- NOTE: This has been generated via update-readme.js -->594595- **browser_navigate_back**596 - Title: Go back597 - Description: Go back to the previous page598 - Parameters: None599 - Read-only: **false**600601<!-- NOTE: This has been generated via update-readme.js -->602603- **browser_network_requests**604 - Title: List network requests605 - Description: Returns all network requests since loading the page606 - Parameters:607 - `includeStatic` (boolean): Whether to include successful static resources like images, fonts, scripts, etc. Defaults to false.608 - `filename` (string, optional): Filename to save the network requests to. If not provided, requests are returned as text.609 - Read-only: **true**610611<!-- NOTE: This has been generated via update-readme.js -->612613- **browser_press_key**614 - Title: Press a key615 - Description: Press a key on the keyboard616 - Parameters:617 - `key` (string): Name of the key to press or a character to generate, such as `ArrowLeft` or `a`618 - Read-only: **false**619620<!-- NOTE: This has been generated via update-readme.js -->621622- **browser_resize**623 - Title: Resize browser window624 - Description: Resize the browser window625 - Parameters:626 - `width` (number): Width of the browser window627 - `height` (number): Height of the browser window628 - Read-only: **false**629630<!-- NOTE: This has been generated via update-readme.js -->631632- **browser_run_code**633 - Title: Run Playwright code634 - Description: Run Playwright code snippet635 - Parameters:636 - `code` (string): 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.getByRole('button', { name: 'Submit' }).click(); return await page.title(); }`637 - `filename` (string, optional): Filename to save the result to. If not provided, result is returned as JSON string.638 - Read-only: **false**639640<!-- NOTE: This has been generated via update-readme.js -->641642- **browser_select_option**643 - Title: Select option644 - Description: Select an option in a dropdown645 - Parameters:646 - `element` (string, optional): Human-readable element description used to obtain permission to interact with the element647 - `ref` (string): Exact target element reference from the page snapshot648 - `values` (array): Array of values to select in the dropdown. This can be a single value or multiple values.649 - Read-only: **false**650651<!-- NOTE: This has been generated via update-readme.js -->652653- **browser_snapshot**654 - Title: Page snapshot655 - Description: Capture accessibility snapshot of the current page, this is better than screenshot656 - Parameters:657 - `filename` (string, optional): Save snapshot to markdown file instead of returning it in the response.658 - Read-only: **true**659660<!-- NOTE: This has been generated via update-readme.js -->661662- **browser_take_screenshot**663 - Title: Take a screenshot664 - Description: Take a screenshot of the current page. You can't perform actions based on the screenshot, use browser_snapshot for actions.665 - Parameters:666 - `type` (string): Image format for the screenshot. Default is png.667 - `filename` (string, optional): 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.668 - `element` (string, optional): Human-readable element description used to obtain permission to screenshot the element. If not provided, the screenshot will be taken of viewport. If element is provided, ref must be provided too.669 - `ref` (string, optional): Exact target element reference from the page snapshot. If not provided, the screenshot will be taken of viewport. If ref is provided, element must be provided too.670 - `fullPage` (boolean, optional): When true, takes a screenshot of the full scrollable page, instead of the currently visible viewport. Cannot be used with element screenshots.671 - Read-only: **true**672673<!-- NOTE: This has been generated via update-readme.js -->674675- **browser_type**676 - Title: Type text677 - Description: Type text into editable element678 - Parameters:679 - `element` (string, optional): Human-readable element description used to obtain permission to interact with the element680 - `ref` (string): Exact target element reference from the page snapshot681 - `text` (string): Text to type into the element682 - `submit` (boolean, optional): Whether to submit entered text (press Enter after)683 - `slowly` (boolean, optional): 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.684 - Read-only: **false**685686<!-- NOTE: This has been generated via update-readme.js -->687688- **browser_wait_for**689 - Title: Wait for690 - Description: Wait for text to appear or disappear or a specified time to pass691 - Parameters:692 - `time` (number, optional): The time to wait in seconds693 - `text` (string, optional): The text to wait for694 - `textGone` (string, optional): The text to wait for to disappear695 - Read-only: **false**696697</details>698699<details>700<summary><b>Tab management</b></summary>701702<!-- NOTE: This has been generated via update-readme.js -->703704- **browser_tabs**705 - Title: Manage tabs706 - Description: List, create, close, or select a browser tab.707 - Parameters:708 - `action` (string): Operation to perform709 - `index` (number, optional): Tab index, used for close/select. If omitted for close, current tab is closed.710 - Read-only: **false**711712</details>713714<details>715<summary><b>Browser installation</b></summary>716717<!-- NOTE: This has been generated via update-readme.js -->718719- **browser_install**720 - Title: Install the browser specified in the config721 - Description: Install the browser specified in the config. Call this if you get an error about the browser not being installed.722 - Parameters: None723 - Read-only: **false**724725</details>726727<details>728<summary><b>Coordinate-based (opt-in via --caps=vision)</b></summary>729730<!-- NOTE: This has been generated via update-readme.js -->731732- **browser_mouse_click_xy**733 - Title: Click734 - Description: Click left mouse button at a given position735 - Parameters:736 - `element` (string): Human-readable element description used to obtain permission to interact with the element737 - `x` (number): X coordinate738 - `y` (number): Y coordinate739 - Read-only: **false**740741<!-- NOTE: This has been generated via update-readme.js -->742743- **browser_mouse_drag_xy**744 - Title: Drag mouse745 - Description: Drag left mouse button to a given position746 - Parameters:747 - `element` (string): Human-readable element description used to obtain permission to interact with the element748 - `startX` (number): Start X coordinate749 - `startY` (number): Start Y coordinate750 - `endX` (number): End X coordinate751 - `endY` (number): End Y coordinate752 - Read-only: **false**753754<!-- NOTE: This has been generated via update-readme.js -->755756- **browser_mouse_move_xy**757 - Title: Move mouse758 - Description: Move mouse to a given position759 - Parameters:760 - `element` (string): Human-readable element description used to obtain permission to interact with the element761 - `x` (number): X coordinate762 - `y` (number): Y coordinate763 - Read-only: **false**764765</details>766767<details>768<summary><b>PDF generation (opt-in via --caps=pdf)</b></summary>769770<!-- NOTE: This has been generated via update-readme.js -->771772- **browser_pdf_save**773 - Title: Save as PDF774 - Description: Save page as PDF775 - Parameters:776 - `filename` (string, optional): File name to save the pdf to. Defaults to `page-{timestamp}.pdf` if not specified. Prefer relative file names to stay within the output directory.777 - Read-only: **true**778779</details>780781<details>782<summary><b>Test assertions (opt-in via --caps=testing)</b></summary>783784<!-- NOTE: This has been generated via update-readme.js -->785786- **browser_generate_locator**787 - Title: Create locator for element788 - Description: Generate locator for the given element to use in tests789 - Parameters:790 - `element` (string, optional): Human-readable element description used to obtain permission to interact with the element791 - `ref` (string): Exact target element reference from the page snapshot792 - Read-only: **true**793794<!-- NOTE: This has been generated via update-readme.js -->795796- **browser_verify_element_visible**797 - Title: Verify element visible798 - Description: Verify element is visible on the page799 - Parameters:800 - `role` (string): ROLE of the element. Can be found in the snapshot like this: `- {ROLE} "Accessible Name":`801 - `accessibleName` (string): ACCESSIBLE_NAME of the element. Can be found in the snapshot like this: `- role "{ACCESSIBLE_NAME}"`802 - Read-only: **false**803804<!-- NOTE: This has been generated via update-readme.js -->805806- **browser_verify_list_visible**807 - Title: Verify list visible808 - Description: Verify list is visible on the page809 - Parameters:810 - `element` (string): Human-readable list description811 - `ref` (string): Exact target element reference that points to the list812 - `items` (array): Items to verify813 - Read-only: **false**814815<!-- NOTE: This has been generated via update-readme.js -->816817- **browser_verify_text_visible**818 - Title: Verify text visible819 - Description: Verify text is visible on the page. Prefer browser_verify_element_visible if possible.820 - Parameters:821 - `text` (string): TEXT to verify. Can be found in the snapshot like this: `- role "Accessible Name": {TEXT}` or like this: `- text: {TEXT}`822 - Read-only: **false**823824<!-- NOTE: This has been generated via update-readme.js -->825826- **browser_verify_value**827 - Title: Verify value828 - Description: Verify element value829 - Parameters:830 - `type` (string): Type of the element831 - `element` (string): Human-readable element description832 - `ref` (string): Exact target element reference that points to the element833 - `value` (string): Value to verify. For checkbox, use "true" or "false".834 - Read-only: **false**835836</details>837838<details>839<summary><b>Tracing (opt-in via --caps=tracing)</b></summary>840841<!-- NOTE: This has been generated via update-readme.js -->842843- **browser_start_tracing**844 - Title: Start tracing845 - Description: Start trace recording846 - Parameters: None847 - Read-only: **true**848849<!-- NOTE: This has been generated via update-readme.js -->850851- **browser_stop_tracing**852 - Title: Stop tracing853 - Description: Stop trace recording854 - Parameters: None855 - Read-only: **true**856857</details>858859860<!--- End of tools generated section -->861
Full transparency — inspect the skill content before installing.