Some links on this page are affiliate links: if you buy through them we may earn a commission, at no extra cost to you.
For a screenshot of a website, use Cursor’s integrated Browser tool: ask Agent to navigate to the page and capture it. If the screenshot must come from your own capture service or tool, configure an MCP server that returns the image as base64 data with an image MIME type. Cursor can attach that returned image to chat for analysis when the selected model supports images. For a screenshot of your actual desktop rather than a web page, use Cursor’s separate computer-use worker workflow.
Choose the screenshot method that matches what you need
“Screenshot with Cursor MCP” can mean capturing a page in Cursor’s controlled browser, passing an image from an external MCP tool into chat, or capturing the computer’s desktop. These are different workflows: MCP is the connection and handoff mechanism for tools, not a universal screenshot engine. Choose the capture source first.
| What you want to capture | Use | What Cursor receives |
|---|---|---|
| A website page in a browser | Cursor’s integrated Browser tool | A screenshot representing the rendered page, available for Agent’s visual feedback |
| An image produced by your service or custom tool | Your MCP server, returning image content | An image attached to chat; analysis depends on whether the selected model supports images |
| Your computer desktop or a locally displayed app | The computer-use worker workflow | A screenshot captured through the worker, subject to platform setup and permissions |
Capture a website with Cursor’s integrated Browser
For an ordinary web-page screenshot, start with Cursor’s integrated Browser rather than building an MCP server. Cursor documents Browser control through MCP tools and describes screenshots as visual representations that help Agent understand layout and confirm browser actions. Ask Agent to open the target page and take a screenshot; then use the image and browser context to discuss layout, visible content, or whether an action produced the expected result.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Scan for outdated or missing drivers - takes under a minute3Repair Windows errors before they cause bigger problemsPractical prompt
Use a prompt that names the destination and the result you want, for example: “Navigate to https://example.com and take a screenshot of the page. Tell me what is visible in the header.” Replace the URL with a page you are authorized to access. If the task concerns a particular state, specify the steps to reach it before asking for the capture, such as opening a menu or navigating to a particular section.
#1 Best Overall
- CRISP CLARITY: This 23.8″ Philips V line monitor delivers crisp Full HD 1920x1080 visuals. Enjoy movies, shows and videos with remarkable detail
- INCREDIBLE CONTRAST: The VA panel produces brighter whites and deeper blacks. You get true-to-life images and more gradients with 16.7 million colors
- THE PERFECT VIEW: The 178/178 degree extra wide viewing angle prevents the shifting of colors when viewed from an offset angle, so you always get consistent colors
- WORK SEAMLESSLY: This sleek monitor is virtually bezel-free on three sides, so the screen looks even bigger for the viewer. This minimalistic design also allows for seamless multi-monitor setups that enhance your workflow and boost productivity
- A BETTER READING EXPERIENCE: For busy office workers, EasyRead mode provides a more paper-like experience for when viewing lengthy documents
What to verify
- Confirm Agent has the Browser capability available and asks it to use the browser for the task.
- Give the page URL explicitly; do not rely on the agent inferring which tab or environment you mean.
- For a stateful page, describe the interaction to perform before the screenshot. A screenshot records the state reached, not the state you intended.
- Inspect the result. A screenshot is a visual representation, so it is useful for layout and visible content but should not be treated as proof that hidden page data or application logic is correct.
Return an image from a custom MCP server
Use a custom MCP tool when the image needs to be created or fetched outside Cursor’s integrated Browser—for example, by a screenshot service or a capture process you operate. Cursor’s MCP documentation describes returning images as base64-encoded data in an image content block with a MIME type. Cursor attaches returned images to chat; whether the model can analyze the image depends on image support in the selected model.
Configure the server and tool
- Add the server through Cursor Marketplace, or configure a custom server in
mcp.json. The exact server configuration and authentication depend on that server; Cursor’s general MCP guidance does not define one universal screenshot server configuration. - Make sure the server exposes a tool that actually obtains or generates the screenshot. Returning an image is the handoff format; it does not itself capture a browser or desktop.
- Have the tool return image content with the image’s base64 data and correct MIME type. Check the capture service’s own documentation for its tool arguments, authentication, and supported image formats.
- In Cursor, verify the server is enabled and its tools are available. Ask Agent to use the specific tool by name, or describe the capture task clearly.
- Check the tool result in chat and confirm the image appears. If the image is present but the model does not discuss its contents, check whether the selected model supports image input.
A custom server is the right choice when capture behavior must come from a particular external system. It adds setup and security responsibilities, though: you must select and configure the server, establish its access to the target, and understand what data it can reach. Cursor advises users to verify MCP server sources, review permissions, use restricted API keys, and audit server code for critical integrations.
Rank #2
- CRISP CLARITY: This 22 inch class (21.5″ viewable) Philips V line monitor delivers crisp Full HD 1920x1080 visuals. Enjoy movies, shows and videos with remarkable detail
- 100HZ FAST REFRESH RATE: 100Hz brings your favorite movies and video games to life. Stream, binge, and play effortlessly
- SMOOTH ACTION WITH ADAPTIVE-SYNC: Adaptive-Sync technology ensures fluid action sequences and rapid response time. Every frame will be rendered smoothly with crystal clarity and without stutter
- INCREDIBLE CONTRAST: The VA panel produces brighter whites and deeper blacks. You get true-to-life images and more gradients with 16.7 million colors
- THE PERFECT VIEW: The 178/178 degree extra wide viewing angle prevents the shifting of colors when viewed from an offset angle, so you always get consistent colors
Use Cursor’s CLI to inspect MCP servers
Cursor says its CLI uses the same MCP configuration as the editor. The documented example for asking the CLI to navigate and take a screenshot is:
agent -p "Navigate to google.com and take a screenshot of the search page"Use your intended destination and task in place of the example. Before troubleshooting a missing tool, inspect the CLI’s configured servers and their tools:
Rank #3
- Clear visuals. Fluid motion: A 144Hz refresh rate and 1ms MPRT deliver smooth, tear‑free motion across work, gaming, and streaming for clearer, more fluid viewing.
- Eye comfort: TÜV Rheinland 3‑star* certification reduces harmful blue light while preserving stunning color quality without compromise. *TÜV Rheinland 3-star eye comfort certification.
- Wide viewing angle: Get consistent views across a wide 178° /178° viewing angle.
- In-Plane Switching (IPS): See excellent color accuracy and consistency across wide viewing angles with In-plane Switching (IPS) technology.
- Ultra-thin bezels: Maximize your viewing experience with thin bezels.
agent mcp list
agent mcp list-tools <server>agent mcp list shows configured servers and status; agent mcp list-tools <server> lists the tools exposed by a named server. These checks help distinguish a server that is not configured or enabled from a prompt that simply did not call the tool you expected.
Capture a self-hosted desktop
If you need the whole desktop or a locally displayed application rather than a page in Cursor’s integrated Browser, use the separate computer-use worker. The setup differs by operating system, and a real capture task is the useful end-to-end verification.
Rank #4
- CURVED FOR ENHANCED ENGAGEMENT: An immersive viewing experience with a curved monitor that wraps more closely around your field of vision; It creates a wider view, enhancing depth perception and minimizing peripheral distraction
- SMOOTH PERFORMANCE FOR SEAMLESS CONTENT: Stay in the action when playing games, watching videos, or working on creative projects; The 100Hz refresh rate reduces lag and motion blur so you don't miss a thing in fast-paced moments¹
- MORE GAMING POWER: Gain the edge with optimizable game settings; Color and image contrast can be adjusted to see scenes more vividly and spot enemies hiding in the dark; Game Mode adjusts any game to fill the screen so you can view every detail²
- KEEP IT EASY ON THE EYES: Care for your eyes and stay comfortable, even during long sessions; Advanced eye comfort technology certified by TÜV reduces eye strain by minimizing blue light and reducing irritating screen flicker²
- INCREASED VERSATILITY: Connect to more; Plug devices straight into your monitor for increased flexibility, making your computing environment even more convenient
macOS
- Start the worker with
agent worker --computer-use start. - Grant Accessibility and Screen Recording permissions to Cursor Computer Use.
- Run a real screenshot task to verify capture.
Cursor’s agent worker debug command confirms helper installation on macOS, but does not report whether Accessibility and Screen Recording permissions have been granted. Grant permissions to Cursor Computer Use—not Terminal or the Cursor desktop app.
Free tools Windows power users keep installed
One-click scans. No signup required.
Linux
Cursor’s computer-use workflow uses X11 and requires the documented desktop packages on Linux. Chrome or Chromium is optional, but recommended for browser use. The worker’s display environment and required packages therefore matter; a configured MCP server alone does not establish that the desktop can be captured.
Best Value
- 【INTEGRATED SPEAKERS】Whether you're at work or in the midst of an intense gaming session, our built-in speakers provide rich and seamless audio, all while keeping your desk clutter-free.
- 【EASY ON THE EYES】 Protect your eyes and enhance your comfort with Blue-Light Shift technology. This feature reduces harmful blue light emissions from your screen, helping to alleviate eye strain during long hours of use and promoting healthier viewing habits.
- 【WIDEN YOUR PERSPECTIVE】Our sleek minimal bezel design ensures undivided attention. The nearly bezel-free display seamlessly connects in a dual monitor arrangement, delivering an unobstructed view that lets you focus on more at once, completely distraction-free.
Account for enterprise browser restrictions
For enterprise Browser access, administrators can enable browser features and configure origin allowlists. When an allowlist is configured, automatic navigation and MCP tool execution are limited to allowed origins. Manual navigation can still display other origins, but browser tools are blocked on those origins. If a Browser task works manually but fails when Agent tries to navigate or invoke a tool, ask the administrator whether the destination is permitted by the organization’s origin policy.
Troubleshoot common screenshot failures
- Agent does not take a screenshot: For a website, ask it directly to use Cursor’s integrated Browser and name the URL. For a custom capture route, confirm the MCP server is enabled and that its screenshot tool is exposed.
- The expected custom tool is missing: In the CLI, run
agent mcp listandagent mcp list-tools <server>. Check server status, configuration, and the tool name the server actually exposes. - The MCP call returns no image: Confirm the tool performs a capture and returns image content containing base64 data and a valid image MIME type. Cursor’s image-return mechanism does not generate the screenshot for the server.
- The image appears but is not interpreted: Image analysis requires a selected model that supports images. Seeing an attachment and having the model analyze its contents are separate outcomes.
- A macOS desktop capture fails despite the helper being installed: Check Accessibility and Screen Recording access for Cursor Computer Use. The debug command does not verify those permission grants.
- A Linux desktop capture fails: Check that the worker has the required documented desktop packages and is using X11. For browser-based desktop work, Chrome or Chromium is recommended.
- An enterprise page is blocked only for automated actions: Check the Browser origin allowlist with the administrator. Manual display of an origin does not mean browser tools are allowed to operate there.
Or skip the browser setup
If you want a website screenshot through an API instead of setting up Cursor’s browser capture, ScreenshotNeo is a website screenshot API and MCP server for developers. One GET request returns an image or PDF. Its clean-shot process accepts consent banners as a visitor and removes more than 60 known consent platforms, newsletter popups, and chat widgets; each step can be turned off. Bot checks, blank pages, timeouts, failed loads, and cache hits cost nothing, with the outcome reported in response headers. Its MCP server exposes take_screenshot, get_page_info, and capture_pdf for AI agents, including Cursor.
curl -G "https://api.screenshotneo.com/v1/shot" -d access_key=YOUR_API_KEY --data-urlencode url=https://stripe.com -o shot.webpSee the ScreenshotNeo API documentation for setup and options. It also supports Python and Node.js requests, along with options such as full-page capture, element selection, viewport and device settings, PDF output, custom CSS and JavaScript, wait conditions, and request blocking. The Free plan includes 1,000 screenshots per month without a card; paid plans start at $5 for 3,000. Learn about ScreenshotNeo, then sign up for 1,000 free screenshots a month with no card.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Security and access checks
- Verify the source of an MCP server before connecting it, and review the permissions it requests.
- Use restricted API keys rather than credentials with broader access than the capture task needs.
- For critical integrations, audit the server code and understand what pages, files, or services it can access.
- For enterprise Browser use, account for administrator-managed origin allowlists as well as server permissions.
Frequently Asked Questions
Does Cursor’s MCP protocol itself take screenshots?
No. The capture comes from Cursor’s Browser, a computer-use worker, or a tool implemented by an MCP server; MCP carries tool interactions and can return image content.
Can I ask Cursor to screenshot a page from the CLI?
Yes. Cursor’s CLI documentation includes the prompt pattern `agent -p “Navigate to google.com and take a screenshot of the search page”`, and CLI MCP shares configuration with the editor.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
