Screenshotline

Official2 tools

by MAMubeenKhan · JavaScript

Screenshot any web page, or read it as clean Markdown, with consent banners, ads and popups removed.

io.github.MAMubeenKhan/screenshotline — MCP Server

This MCP server captures screenshots of web pages and can also read web content as clean Markdown. It targets cleaner output by removing consent banners, advertisements, and popups, and supports representing page content either visually (screenshots) or as text (Markdown).

🛠️ Key Features

Screenshot any web page
Read web page content as clean Markdown
Removes consent banners, ads, and popups

🚀 Use Cases

Generate screenshots for web pages with fewer overlays
Extract page content into Markdown without common obstructing elements

⚡ Developer Benefits

Cleaner Markdown suitable for downstream processing
Screenshot outputs for documentation or verification with reduced UI noise

⚠️ Limitations

Capabilities beyond screenshots/Markdown cleaning are not specified in the provided server data

Topics

cookie-bannerdockerheadless-chromehtml-to-imagemcp-servermodel-context-protocolnodejspuppeteerscreenshotscreenshot-apiself-hostedurl-to-imageweb-to-markdownwebsite-screenshot

Captured live from the server via tools/list.

screenshot

Capture a screenshot of a web page and return it as an image. Handles consent banners, ads, lazy-loaded images and late-rendering SPAs. Use this when the layout, styling or visual appearance is what matters; use read_page when you only need the text.

Parameters11

url	string	required	The page to capture. Must be http or https.
full_page	boolean	optional	Capture the entire scrollable page rather than just the viewport.
viewport_width	number	optional	Viewport width in pixels. Default 1280.
viewport_height	number	optional	Viewport height in pixels. Default 800.
format	string	optional	Default png.
selector	string	optional	Capture only the element matching this CSS selector.
color_scheme	string	optional	Render the page in light or dark mode.
block_ads	boolean	optional	Block ad networks and collapse the empty slots. Default true.
block_cookie_banners	boolean	optional	Remove consent banners. Default true.
block_chats	boolean	optional	Remove live-chat widgets. Default true.
block_popups	boolean	optional	Remove newsletter and donation interstitials. Default true.

Raw schema

{
  "type": "object",
  "properties": {
    "url": {
      "type": "string",
      "description": "The page to capture. Must be http or https."
    },
    "full_page": {
      "type": "boolean",
      "description": "Capture the entire scrollable page rather than just the viewport."
    },
    "viewport_width": {
      "type": "number",
      "description": "Viewport width in pixels. Default 1280."
    },
    "viewport_height": {
      "type": "number",
      "description": "Viewport height in pixels. Default 800."
    },
    "format": {
      "type": "string",
      "enum": [
        "png",
        "jpeg",
        "webp"
      ],
      "description": "Default png."
    },
    "selector": {
      "type": "string",
      "description": "Capture only the element matching this CSS selector."
    },
    "color_scheme": {
      "type": "string",
      "enum": [
        "light",
        "dark"
      ],
      "description": "Render the page in light or dark mode."
    },
    "block_ads": {
      "type": "boolean",
      "description": "Block ad networks and collapse the empty slots. Default true."
    },
    "block_cookie_banners": {
      "type": "boolean",
      "description": "Remove consent banners. Default true."
    },
    "block_chats": {
      "type": "boolean",
      "description": "Remove live-chat widgets. Default true."
    },
    "block_popups": {
      "type": "boolean",
      "description": "Remove newsletter and donation interstitials. Default true."
    }
  },
  "required": [
    "url"
  ]
}

read_page

Render a web page and return its content as clean Markdown. The page is fully rendered first - JavaScript executed, consent banners dismissed, lazy images loaded - so this works on sites that return nothing useful to a plain HTTP fetch. Prefer this over screenshot when you need to read or reason about the text.

Parameters6

url	string	required	The page to read. Must be http or https.
max_chars	number	optional	Truncate the Markdown to this many characters. 0 or omitted means no limit.
strip_chrome	boolean	optional	Drop navigation, headers, footers and sidebars, keeping the main content. Default true. Set false for a full transcript of the page.
block_cookie_banners	boolean	optional	Remove consent banners. Default true.
block_chats	boolean	optional	Remove live-chat widgets. Default true.
block_popups	boolean	optional	Remove newsletter and donation interstitials. Default true.

Raw schema

{
  "type": "object",
  "properties": {
    "url": {
      "type": "string",
      "description": "The page to read. Must be http or https."
    },
    "max_chars": {
      "type": "number",
      "description": "Truncate the Markdown to this many characters. 0 or omitted means no limit."
    },
    "strip_chrome": {
      "type": "boolean",
      "description": "Drop navigation, headers, footers and sidebars, keeping the main content. Default true. Set false for a full transcript of the page."
    },
    "block_cookie_banners": {
      "type": "boolean",
      "description": "Remove consent banners. Default true."
    },
    "block_chats": {
      "type": "boolean",
      "description": "Remove live-chat widgets. Default true."
    },
    "block_popups": {
      "type": "boolean",
      "description": "Remove newsletter and donation interstitials. Default true."
    }
  },
  "required": [
    "url"
  ]
}

mcp/README.md2 KBView on GitHub

screenshotline-mcp

An MCP server that gives an agent two tools:

screenshot — capture a web page as an image, for when the layout is the answer.
read_page — render the page, then return its text as clean Markdown, for when the text is.

Pages are fully rendered first: JavaScript executed, consent banners removed, lazy images loaded, ads, chat widgets and newsletter popups stripped. So read_page works on sites that return nothing useful to a plain HTTP fetch. A capture that comes back blank is reported as a warning, not narrated as though it were the page.

Install

Get a free key at screenshotline.com/account — 500 renders a month, no card. Then, in Claude Code:

bash

claude mcp add screenshotline -e SCREENSHOTLINE_ACCESS_KEY=sl_live_... -- npx -y screenshotline-mcp

Or in any MCP client's JSON config (Claude Desktop, Cursor, and others):

json

{
  "mcpServers": {
    "screenshotline": {
      "command": "npx",
      "args": ["-y", "screenshotline-mcp"],
      "env": { "SCREENSHOTLINE_ACCESS_KEY": "sl_live_..." }
    }
  }
}

Self-hosted

The renderer is open source. Point the server at your own instance and no key is needed unless you configured one:

bash

SCREENSHOTLINE_BASE_URL=http://localhost:3000 npx -y screenshotline-mcp

Variable	Default
`SCREENSHOTLINE_ACCESS_KEY`	—	Your API key. Required for the hosted API.
`SCREENSHOTLINE_BASE_URL`	`https://api.screenshotline.com`	Any Screenshotline instance.

This package is only the MCP bridge — one file plus the MCP SDK. It does not download a browser; the rendering happens on the API it points at.

Docs: screenshotline.com/docs/extract · License: AGPL-3.0-or-later

Install

Tools (2)

screenshotread_page

Remote endpoint

Streamable HTTP

Hosted server - connect over the network, no local install.

url

https://api.screenshotline.com/mcp

Add to VS Code Add to Cursor

claude_desktop_config.json

json

{
  "mcpServers": {
    "screenshotline": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://api.screenshotline.com/mcp"
      ]
    }
  }
}

Desktop config is stdio-only; this bridges via mcp-remote. Native remote: Settings > Connectors.

io.github.MAMubeenKhan/screenshotline — MCP Server

🛠️ Key Features

🚀 Use Cases

⚡ Developer Benefits

⚠️ Limitations

Topics

screenshot

read_page

screenshotline-mcp

Install

Self-hosted

Install

Remote endpoint

Related servers