Give Your AI Assistant Full PDF Superpowers
Connect Claude, ChatGPT, Cursor, or any MCP-compatible AI to 25 PDF tools. Merge, split, compress, redact, sign, check accessibility — through conversation instead of code.
Works with every AI assistant that supports the Model Context Protocol.
One Connection, 25 Tools, Zero Code
PDFlys MCP is a Model Context Protocol server that gives your AI assistant access to 25 PDF tools. Describe what you need in plain language — instead of writing integration code or juggling separate apps.
Your AI connects to PDFlys MCP once. After that, you just ask for what you need.
Need to merge 12 contracts into one file? Ask your AI.
Need to redact social security numbers from 50 pages? Ask your AI.
Need to check a government form for WCAG 2.1 accessibility compliance? Ask your AI.
File transfer, processing, error recovery, and result delivery are all handled server-side. Your AI picks the right tool and gets you the file.
Connect
Add one config line to your AI assistant
Ask
Describe what you need in plain language
Done
Get your processed PDF back instantly
Why Developers and Teams Are Choosing MCP Over Traditional APIs
| Traditional PDF API | PDFlys MCP Server | |
|---|---|---|
| Setup | Read docs, get API keys, write integration code, handle errors | Add one config line, start asking |
| Learning curve | Study endpoints, parameters, authentication flows | Your AI already knows every tool |
| Error handling | Write try/catch blocks, parse error codes, implement retries | AI reads the error and adjusts automatically |
| Multi-step workflows | Chain API calls manually, manage intermediate files | "Compress, watermark, add page numbers" — one sentence |
| Maintenance | Update SDK versions, fix breaking changes | We update the server; your setup stays the same |
| Who can use it | Developers only | Anyone who can type a sentence |
Ask your AI to merge two contracts. Without MCP, it gives you instructions. With MCP, it hands you the merged file.
25 PDF Tools, Ready When Your AI Is
Every tool that powers the PDFlys web editor — now available to your AI assistant through a single MCP connection.
Document Assembly
Merge PDFs
Combine multiple PDFs into a single file. Supports 2+ files in one operation.
Split PDF
Divide a PDF by page ranges, every page, or every N pages.
Extract Pages
Pull specific pages into a new PDF.
Remove Pages
Delete unwanted pages from a document.
Organize Pages
Reorder pages by specifying the exact sequence you want.
Document Enhancement
Add Watermark
Apply text or image watermarks with control over position, opacity, rotation, and color.
Add Page Numbers
Insert page numbers in any position with customizable format, font, and color.
Add Text
Place text annotations anywhere on any page with font, size, and color options.
Add Shapes
Draw rectangles, circles, ellipses, and lines with fill, stroke, and opacity controls.
Highlight PDF
Add colored highlight overlays to rectangular regions on any page.
Sign PDF
Overlay signature images with optional date stamps and additional text.
Security & Compliance
Protect PDF
Apply password encryption with separate user and owner passwords.
Unlock PDF
Remove password protection from a secured PDF.
Redact PDF
Permanently black out sensitive regions — coordinates-based precision redaction.
Accessibility Check
Run a WCAG 2.1, Section 508, PDF/UA, and EN 301 549 compliance audit with scored reports.
Export Report (PDF)
Download a full PDF accessibility audit report for compliance records.
Export Report (CSV)
Export accessibility findings as CSV for spreadsheet analysis and tracking.
Conversion & Optimization
Compress PDF
Reduce file size with three compression levels or fine-tune with custom quality and DPI settings.
PDF to Image
Convert pages to JPG, PNG, or WebP at configurable DPI and quality.
PDF to JPG
Quick conversion of PDF pages to high-quality JPG images.
Image to PDF
Convert one or more images into a PDF with page size, orientation, and margin control.
Analysis & Comparison
Compare PDFs
Compare two PDFs with text-level, visual, or combined analysis. Returns per-page differences.
File Management
Upload Files
Get secure presigned upload URLs for up to 10 files at once. Supports PDF, PNG, JPG, and WebP.
Health Check
Verify API status, current usage, plan limits, and reset dates.
Chain Operations Like a Pipeline — In Plain English
Most real-world PDF tasks aren't a single operation. You need to compress, then watermark, then add page numbers, then password-protect. With traditional tools, that means four separate steps with four file handoffs.
With PDFlys MCP, you describe the entire pipeline in one request:
Compressed: 50MB → 12MB (high compression)
Watermark added: "CONFIDENTIAL" diagonal, 30% opacity
Page numbers added: bottom center, "Page {n} of {total}"
Password protection applied
Here's your processed file: contract_final.pdfHow it works under the hood: Each tool returns a fileId that the next tool accepts as input. Files stay server-side between steps, so nothing gets re-uploaded or re-encoded. Your AI manages the chain.
How Teams Use PDFlys MCP Every Day
Legal — Contract Assembly and Redaction
A legal team receives signed contract pages, appendices, and exhibits as separate files. Before sharing with opposing counsel, they need to combine everything, redact privileged information, and add page numbers.
"Merge these 7 PDF files in order. Then redact the social security numbers on pages 12 and 15. Add page numbers at the bottom center in 'Page X of Y' format."
Compliance — Accessibility Auditing at Scale
A government agency must ensure all public-facing PDFs meet WCAG 2.1 AA and Section 508 standards. Manual auditing takes hours per document.
"Run an accessibility check on this PDF against WCAG 2.1 AA and Section 508 standards. Then export the results as both a PDF report and a CSV."
Marketing — Brand-Consistent Document Production
A marketing team produces whitepapers weekly. Every document needs the company watermark, consistent page numbering, optimized file size, and password protection for pre-release drafts.
"For each of these 4 whitepapers: compress to medium quality, add our logo as a watermark at 20% opacity, add page numbers, and protect with password."
Finance — Invoice Processing and Archival
An accounting team receives invoices as images and needs to convert them to PDF, combine into monthly bundles, compress for storage, and add text labels.
"Convert these 15 invoice images to PDF. Merge them into one file. Add the text 'Q1 2026 — Vendor Invoices' at the top. Compress the final file."
Education — Exam Preparation and Distribution
A professor builds exams by extracting questions from a test bank PDF, reordering them, adding a watermark, and creating print and secure digital versions.
"Extract pages 5, 12, 18, 23, and 31 from the test bank. Reorder so page 31 comes first. Add a 'DO NOT DISTRIBUTE' watermark. Create a regular and a password-protected version."
Real Estate — Document Comparison and Review
A real estate firm receives updated lease agreements and needs to identify exactly what changed between the previous version and the new draft.
"Compare these two lease agreements. Show me both text changes and visual differences page by page."
HR — Onboarding Packet Assembly
An HR department needs to compile personalized onboarding packets — combining handbook, benefits guide, tax forms, and welcome letter — then sign, number, and deliver.
"Merge the handbook, benefits guide, W-4 form, and welcome letter. Add page numbers starting from page 2. Add my signature to page 45. Compress for email."
Set Up in 60 Seconds
Add PDFlys MCP to any compatible AI assistant. One config file, no code.
- Open your Claude Desktop config file
- Add the PDFlys MCP server configuration
- Restart Claude Desktop — you'll be prompted to authenticate via OAuth
{
"mcpServers": {
"pdflys": {
"url": "https://mcp.pdflys.com/mcp"
}
}
}Getting Your API Key
- Go to pdflys.com and create a free account
- Navigate to Settings → API Keys
- Create a new key and copy it
- Paste it into the config above
Security Built for Regulated Industries
Secure File Transfer
Every file upload uses presigned URLs with automatic expiration. Files are transferred directly to encrypted cloud storage — never through intermediate servers.
OAuth 2.1 Authentication
Full OAuth 2.1 resource server implementation with JWKS key validation, token revocation via Redis, and session-scoped access.
Session Isolation
Each MCP session gets its own API client, file namespace, and state. Nothing is shared between sessions. Idle sessions expire after 30 minutes.
Built for Production
The Only MCP Server with Built-In Accessibility Auditing
Government agencies, schools, and public-facing organizations are legally required to publish accessible documents. PDFlys MCP is the only MCP server that includes compliance checking out of the box:
Four-Standard Audit
Run your PDF against four international standards simultaneously — WCAG 2.1 AA, Section 508, PDF/UA 1, and EN 301 549. Get a scored result (0–100) with categorized findings.
Detailed Evidence
Every finding includes the specific element that failed, the page it appears on, and which standard clause it violates.
Export for Compliance Records
Generate a professional PDF audit report or a CSV export for tracking in spreadsheets and compliance management systems.
AI-Powered Remediation Guidance
Because results flow directly into your AI conversation, your assistant can immediately explain findings and suggest specific remediation steps.
Frequently Asked Questions
What is MCP and why does it matter for PDF tools?
Which AI assistants work with PDFlys MCP?
Do I need to be a developer to use this?
How is this different from the PDFlys web editor?
What happens to my files?
Is there a file size limit?
Can I chain multiple operations together?
What accessibility standards does the checker support?
Is PDFlys MCP free?
Start Processing PDFs with AI
Add PDFlys MCP to your AI assistant in 60 seconds. Free tier included, no credit card required.