Remote MCP server now live

The HTML-to-PDF API
built for AI agents

Give Claude, Codex, Cursor, OpenCode, and any MCP client a production PDF engine. Generate pixel-perfect PDFs from agent-authored HTML through MCP or REST.

Connect Your AI AgentGet a Free API Key
75 free credits/month
MCP + REST API
MCP-native PDF generation

Give your AI agent a PDF engine

Connect once through Model Context Protocol. Claude, Codex, Cursor, OpenCode, and other MCP clients can turn their own HTML into production-ready PDFs without custom glue code.

Explore the MCP serverGet an API key

pixeltopdf / MCP

Streamable HTTP · authenticated · agent-ready

Streamable HTTP
tool generate_pdf

HTML or URL → embedded PDF resource

tool capture_screenshot

HTML or URL → PNG, JPEG, or WebP

prompt create_pdf_document

Reusable workflow for accessible, print-ready documents

Claude Code

Codex

Cursor

OpenCode v2

VS Code

Prompts never contain your key. Configure it privately in your client, never in agent chat or source control.

🤖Native MCP Server
✨AI Agent Ready
📄HTML to PDF
📸Screenshots
⚡REST API
🌐Chromium Rendering

The complete PDF API for AI agents

Let your agent write the HTML and PixelToPdf handle production rendering. Use native MCP tools for conversational workflows or the REST API for application code.

Connect with MCPView Guides

Built for developers who care

We obsess over the details so you can focus on building great products

We care about predictable output

Control when rendering happens

Wait for network activity, load deferred images, add a bounded delay, and set an explicit timeout for each conversion.

Network idle optionLazy image loadingTimeout controls

We care about your needs

Full control over every detail

From custom headers and footers to CSS injection and JavaScript execution. Support for Tailwind, Bootstrap, and any CSS framework.

CSS/JS injectionCustom headers & footersPage breaks control

We care about your security

Keep conversion keys server-side

Requests use API key authentication over HTTPS. Generated files follow the deletion policy described in our privacy documentation.

API key authenticationHTTPS transportDocument deletion policy

We care about your privacy

Clear data processing terms

PixelToPdf is operated by Dinacode SL in Spain. Our public privacy and GDPR pages explain document processing, retention, and user rights.

EU-based operatorPublished retention policyGDPR rights information

Built for AI-generated documents

MCP-native workflows backed by a high-fidelity Chromium rendering API

Native MCP Server

Give any compatible AI agent authenticated PDF and screenshot tools over Streamable HTTP.

Agent-Ready Output

Return PDFs as embedded MCP resources and screenshots as image content in the same conversation.

Multiple Formats

PDF, PNG, JPEG, or WebP. Choose the format that fits your needs.

URL & HTML

Convert any URL or raw HTML content with a single API call.

Custom Styling

Inject CSS and JavaScript. Support for Tailwind, Bootstrap, and more.

Headers & Footers

Add page numbers, dates, or custom HTML to every page.

Load Control

Wait for network activity, lazy-loaded images, or a custom delay before capture.

Page Controls

Set paper format, orientation, margins, page ranges, print mode, and zoom.

Element Screenshots

Capture a full page or target one element with a CSS selector.

Output Control

Return binary output or Base64 and set a download filename in the request.

Render Safety

Disable JavaScript when it is not needed and set explicit conversion timeouts.

Production use cases

From HTML to a finished document

Follow practical patterns for the documents and screenshots developers generate most often.

Explore all use cases →
01

Invoice PDF API

Generate consistent invoices from your own HTML with print CSS, page numbers, and controlled filenames.

02

Webpage to PDF

Capture public web pages after JavaScript, fonts, images, and network requests have finished loading.

03

PDF reports

Turn dashboards and recurring business reports into paginated, downloadable documents.

04

Website screenshots

Create full-page or element-level PNG, JPEG, and WebP captures through the same REST API.

Connect an agent in minutes

Build your first AI-generated PDF with MCP, Guides, or the Playground

MCP for AI Agents

Connect Claude, Codex, Cursor, and OpenCode

Integration Guides

Step-by-step tutorials for every language

API Playground

Test the API directly in your browser

Pricing

Simple, transparent pricing plans

MCP first, REST when you need it

Connect an AI client or use a guide for application-level integration

Node.jsPythonPHPGo guideRubyJavaC#cURL

Simple, transparent pricing

Start free, scale as you grow. No hidden fees.

Free

2MB max file size

€0forever

75 credits/mo

30s timeout

    Start Free

    Startup

    100MB max file size

    €10/mo

    750 credits/mo

    €0.013/extra credit

    • Templates
    • 2 team seats
    Get Started
    Most Popular

    Boost

    100MB max file size

    €25/mo

    3.8k credits/mo

    €6.67/1k/extra credit

    • Templates
    • Webhooks
    • Async processing
    • 3 team seats
    Get Started

    Growth

    100MB max file size

    €40/mo

    7.5k credits/mo

    €5.33/1k/extra credit

    • Templates
    • Webhooks
    • Async processing
    • 5 team seats
    Get Started

    Business

    100MB max file size

    €100/mo

    37.5k credits/mo

    €2.67/1k/extra credit

    • Templates
    • Webhooks
    • Async processing
    • Custom storage
    Get Started

    Need more? Check out our high-volume plans.

    View all plans

    Frequently asked questions

    Everything you need to know about PixelToPdf

    Can an AI agent generate PDFs directly?
    Yes. PixelToPdf is exposed as a remote MCP server with generate_pdf and capture_screenshot tools for Claude, Codex, Cursor, OpenCode, and other MCP clients.
    Do I need to write integration code for MCP?
    No. Create an API key, copy the setup prompt for your AI client, and connect to the remote MCP endpoint. The same key, credits, and rate limits work across MCP and REST.
    What counts as a conversion?
    Each API call that generates a PDF or image counts as one conversion. Failed requests are not counted.
    Can I convert dynamic websites?
    Yes. Chromium executes page JavaScript, and you can wait for network activity to settle or add a render delay before generating the output.
    What happens if I exceed my monthly limit?
    Your conversions will be paused until the next billing cycle. You can upgrade your plan at any time or enable overage billing.
    Do you support JavaScript-heavy websites?
    Yes. We use a full Chromium browser that executes JavaScript. Use wait_for_network, lazy_load_images, or a delay to control when the output is captured.
    How long are generated files stored?
    Generated files are automatically deleted within one hour after conversion. We do not retain them longer than necessary to deliver the service.
    Can I use custom fonts?
    Yes. You can use web fonts via CSS or embed fonts directly in your HTML. We support all Google Fonts out of the box.
    Is there a file size limit?
    The Free plan supports files up to 2MB. Paid plans support files up to 100MB. Large files may take longer to process.
    Do you offer refunds?
    Subscriptions can be canceled at any time and remain active until the end of the billing period. Refunds are considered case by case, as described in our Terms of Service.

    Give your agent a PDF engine

    Create a key, copy one setup prompt, and start generating PDFs through MCP. You get 75 free conversion credits every month.

    Get a Free API KeyView MCP Setup
    No credit card
    75 free credits/month
    Cancel anytime