Hosted MCP server: render any URL to print-stable PDF, PNG, WebP or Markdown
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
webcapture API key as Bearer token (free tier: 50 renders/month, webcapture.io/docs)
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
List the pricing tiers with their monthly price in USD and their included monthly conversions.
This tool takes no arguments.
This tool isn't marked read-only, so it may change something on the server or in an account you connected.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
claude mcp add --transport http webcapture https://webcapture.io/mcp \ --header 'Authorization: Bearer YOUR_TOKEN'
Headers
Authorization (required): webcapture API key as Bearer token (free tier: 50 renders/month, webcapture.io/docs)List the pricing tiers with their monthly price in USD and their included monthly conversions.
No arguments.
{
"type": "object",
"properties": {}
}Report the calling key's monthly quota state: tier, calendar month, renders used, monthly limit, remaining renders, and the date the count resets.
No arguments.
{
"type": "object",
"properties": {}
}Render a web page to PDF, PNG, JPEG, WebP or Markdown. Returns the download URL of the finished file; fetch it with your API key as the Bearer token.
| Argument | Type | Description |
|---|---|---|
| url* | string | Absolute http(s) URL of the page to render |
| format | string | Output format, defaults to pdf |
| fullPage | boolean | Capture the full scrollable page. Defaults to true; set false for a viewport-only capture. Applies to png, jpeg and webp |
| viewport | object | Layout viewport in pixels before capture, 200 to 7680 each. Applies to png, jpeg and webp |
| landscape | boolean | Landscape orientation, defaults to false |
| colorScheme | string | Render the page as if the visitor prefers light or dark mode; site keeps the page's own choice. Defaults to light |
{
"type": "object",
"required": [
"url"
],
"properties": {
"url": {
"type": "string",
"description": "Absolute http(s) URL of the page to render"
},
"format": {
"enum": [
"pdf",
"png",
"jpeg",
"webp",
"markdown"
],
"type": "string",
"description": "Output format, defaults to pdf"
},
"fullPage": {
"type": "boolean",
"description": "Capture the full scrollable page. Defaults to true; set false for a viewport-only capture. Applies to png, jpeg and webp"
},
"viewport": {
"type": "object",
"required": [
"width",
"height"
],
"properties": {
"width": {
"type": "integer"
},
"height": {
"type": "integer"
}
},
"description": "Layout viewport in pixels before capture, 200 to 7680 each. Applies to png, jpeg and webp"
},
"landscape": {
"type": "boolean",
"description": "Landscape orientation, defaults to false"
},
"colorScheme": {
"enum": [
"light",
"dark",
"site"
],
"type": "string",
"description": "Render the page as if the visitor prefers light or dark mode; site keeps the page's own choice. Defaults to light"
}
}
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Oct 30, 2026.
No readOnlyHint / destructiveHint annotations, so clients can't tell a read from a write without asking.
Fix: Add annotations to each tool, e.g. { readOnlyHint: true } for tools that only read.