CaptureKit MCP Server
OfficialUse CaptureKit through MCP to analyze, capture, extract, crawl, and inspect webpages plus account/infrastructure usage.
Analyze a webpage with AI to get insights, value proposition, features, categories, use cases, and optional contact signals.
Capture webpages as PNG, JPEG, WebP, or PDF with device emulation, full-page capture, selectors, lazy-load scrolling, ad/cookie-banner removal, proxies, caching, and S3 storage options.
Extract content such as structured data, metadata, links, HTML, Markdown, and sitemaps, with cleaning via Defuddle, resource/URL blocking, delays, and selectors.
Crawl pages in request mode without rendering to fetch raw HTML for scraping, audits, monitoring, and link/metadata extraction.
Check stats for infrastructure health of an engine or endpoint over the last hour, if enabled for the account.
Check usage for real-time account statistics and credit balance.
Run locally over stdio or as a Streamable HTTP server/hosted endpoint using a CaptureKit API key; API-calling tools consume credits.
Supports optionally uploading captured screenshots/files to Cloudflare R2 as S3-compatible storage via the capture tool's S3 storage settings.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@CaptureKit MCP Server screenshot https://github.com as a PNG"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
CaptureKit MCP Server
Official Model Context Protocol server for the CaptureKit screenshot API.
Capture websites as PNG, JPEG, WebP or PDF, extract their content, or analyse them with AI.
It lets Claude, Cursor, VS Code, Windsurf and any MCP client call the API as tools. Runs over stdio (local) or Streamable HTTP (self-hosted).
Requirements
Related MCP server: webcap
Hosted server (recommended)
CaptureKit runs this server for you at https://mcp.capturekit.dev/ (Streamable HTTP). Send your API key in the x-api-key header; nothing to install.
claude mcp add --transport http capturekit https://mcp.capturekit.dev/ --header "x-api-key: YOUR_API_KEY"{
"mcpServers": {
"capturekit": {
"url": "https://mcp.capturekit.dev/",
"headers": {
"x-api-key": "YOUR_API_KEY"
}
}
}
}The hosted server and this package expose the same API; run the package locally when you want stdio, your own network egress, or to self-host it.
Run it locally
Claude Code
claude mcp add capturekit --env CAPTUREKIT_API_KEY=YOUR_API_KEY -- npx -y @capturekitdev/mcpClaude Desktop, Cursor, Windsurf and other clients
Add the server to the client's MCP configuration (claude_desktop_config.json, .cursor/mcp.json…):
{
"mcpServers": {
"capturekit": {
"command": "npx",
"args": ["-y", "@capturekitdev/mcp"],
"env": {
"CAPTUREKIT_API_KEY": "YOUR_API_KEY"
}
}
}
}VS Code
{
"servers": {
"capturekit": {
"type": "stdio",
"command": "npx",
"args": ["-y", "@capturekitdev/mcp"],
"env": {
"CAPTUREKIT_API_KEY": "${input:capturekit-api-key}"
}
}
},
"inputs": [
{
"type": "promptString",
"id": "capturekit-api-key",
"description": "CaptureKit API key",
"password": true
}
]
}Claude Desktop extension
Download the .mcpb bundle from the releases page and open it; Claude Desktop asks for your API key.
Docker
docker build -t capturekit-mcp .
docker run -i --rm -e CAPTUREKIT_API_KEY=YOUR_API_KEY capturekit-mcpTools
Tool | Description |
| AI Content API analyzes any webpage and returns AI insights summary, value proposition, features, categories, use cases, and optional contact signals. |
| Browser-like Screenshot API that captures webpages as PNG/JPEG/WebP or PDF for previews, QA, reporting, monitoring, and archiving no headless setup. |
| CaptureKit Content API extracts structured webpage data, metadata, links, HTML/Markdown: in one request for SEO, monitoring, indexing and automation. |
| Infrastructure health for an engine or endpoint over the last hour. Not scoped to your account or tenant. Enabled per account: HTTP 403 means contact support to activate it. |
| Retrieve real-time usage statistics and credit balance for your CaptureKit account. |
| Website Crawler API fetches raw HTML from any webpage in request mode (no rendering) for fast scraping, audits, monitoring, and link/metadata extraction. |
Every tool that calls the API consumes credits. JSON results are returned as text; images are returned inline and other files (PDF…) as embedded resources, up to 4 MB.
Streamable HTTP
npx @capturekitdev/mcp --http --port 3000The endpoint is POST /mcp (stateless, JSON responses). Each client sends its own key in the x-api-key header (or Authorization: Bearer <key>); the server never uses a key of its own. It binds to 127.0.0.1 by default and rejects browser origins unless you allow them — put it behind TLS before exposing it.
createHttpHandler() is exported for Fetch-compatible runtimes (Cloudflare Workers, Deno, Bun):
import { createHttpHandler } from "@capturekitdev/mcp";
export default { fetch: createHttpHandler({ allowedOrigins: [] }) };Security
The API key is read from
CAPTUREKIT_API_KEY(stdio) or the request headers (HTTP) and removed from error messages.Content fetched from websites is untrusted: the server instructions tell the model to treat it as data, not instructions.
See SECURITY.md to report a vulnerability.
Development
npm install
npm test
npm run build
npx @modelcontextprotocol/inspector node dist/stdio.jsBuilt on the SDK
The server uses the official Node.js SDK (@capturekitdev/sdk).
All CaptureKit SDKs
SDK | Package | Install |
Node.js |
| |
Python |
| |
Go |
| |
.NET |
| |
Ruby |
| |
PHP |
| |
Java | build from source, see its README | |
CLI |
| |
MCP server |
|
License
MIT — see LICENSE.
Available Tools
6 toolsanalyzeAnalyze (AI)ARead-onlyInspect
Analyze (AI). AI Content API analyzes any webpage and returns AI insights summary, value proposition, features, categories, use cases, and optional contact signals. Credit cost: 1 credit per call.
Inputs: url (required), cache, proxy, fields, +3 more.
captureCapture (Screenshot)BRead-onlyInspect
Capture (Screenshot). Browser-like Screenshot API that captures webpages as PNG/JPEG/WebP or PDF for previews, QA, reporting, monitoring, and archiving no headless setup. Credit cost: 1 credit per call.
Inputs: url (required), cache, delay, proxy, +26 more.
contentContentBRead-onlyInspect
Content. CaptureKit Content API extracts structured webpage data, metadata, links, HTML/Markdown: in one request for SEO, monitoring, indexing and automation. Credit cost: 1 credit per call.
Inputs: url (required), cache, delay, proxy, +13 more.
statsStatsARead-onlyInspect
Stats. Infrastructure health for an engine or endpoint over the last hour. Not scoped to your account or tenant. Enabled per account: HTTP 403 means contact support to activate it.
Inputs: path, engine.
usageUsageBRead-onlyInspect
Usage. Retrieve real-time usage statistics and credit balance for your CaptureKit account.
Inputs: no named parameters.
website_crawlerWebsite CrawlerARead-onlyInspect
Website Crawler. Website Crawler API fetches raw HTML from any webpage in request mode (no rendering) for fast scraping, audits, monitoring, and link/metadata extraction. Credit cost: 1 credit per call.
Inputs: query (required), fields, allow_redirects, return_page_source.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 6 tools
The screenshot, AI-analyze, and account-usage tools are clearly distinct, but content and website_crawler both advertise HTML/Markdown/link/metadata extraction, making selection between them non-obvious without reading closely. Usage and stats are also easily confused at a glance since neither name signals that one is account-scoped billing and the other is anonymous infrastructure health.
All names are lowercase single tokens except website_crawler, which introduces an underscore compound, so the pattern is readable but not uniform. More importantly, the names are bare nouns with no verb or action cue, so the naming scheme conveys resource but not operation.
Six tools map cleanly onto a focused API surface (render, extract, crawl, analyze, account usage, infra health) with no filler tools. This is a well-scoped set for an API-wrapper server.
The main output modalities (image/PDF, structured content, raw HTML, AI insights) plus account and health introspection are all covered. Minor gaps exist around batch/multi-URL requests, async job status, and management of stored captures, but agents can work around these.
Maintenance
Related MCP Connectors
Web scraping, browser automation, screenshots and AI data extraction for AI agents via MCP.
Web scraping, extraction and search for AI agents. Hosted MCP with workspace API keys.
MCP server for web extraction and rendering via AceDataCloud WebExtrator
Screenshot, PDF, OG-image, and page extraction (markdown/JSON) over MCP. Bearer key or x402.
Related MCP Servers
AlicenseAqualityBmaintenanceEU-native web scraping for AI agents & developers. scrape web pages into markdown, screenshots, metadata, and links, map sites, run asynchronous scrape jobs, and check usage, with EU-hosted, GDPR-aligned processing.7257 npm1Apache 2.0- AlicenseNot gradedqualityBmaintenanceEnables screenshotting any URL, extracting structured page data, and setting up page monitoring alerts via pay-per-call MCP tools.MIT

MeshArc MCP Serverofficial
AlicenseNot gradedqualityCmaintenanceEnables MCP clients to scrape and crawl websites into clean markdown, map site URLs, run browser interactions, extract structured fields, and monitor projects for page changes over time. It also supports searching stored crawl data and following long-running jobs.MIT- FlicenseNot gradedqualityAmaintenanceEnables MCP clients to scrape web pages, extract structured data, search, crawl sites, generate schemas, schedule monitors, and inspect credits/history through hosted or local integrations.-