Paperless-ngx MCP. Your AI agent becomes a digital librarian for your documents.
Paperless-ngx connects your digital archive to any AI agent, turning massive document folders into a searchable knowledge base. Use it to find tax receipts, contract details, or research papers instantly by asking natural questions. Manage tags, upload new files, and retrieve OCR text directly from your existing records.
Give Claude and any AI agent real-world access
Find specific documents by querying the full text, date ranges, or predefined organizational tags.
Upload new files to the archive, download originals for review, or delete outdated records.
Create, update, or retrieve tags and document types to keep your entire library consistently organized.
Fetch the complete OCR text and detailed metadata for any file so your agent can analyze its contents fully.
Ask an AI about this
Waiting for input…
What AI agents can do with Paperless-ngx: 26 Tools for Document Management
These tools give your agent the specific actions it needs to manage documents, metadata, correspondents, and tags within your Paperless-ngx archive.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Paperless-ngx MCPList Tags
Provides a complete list of every tag currently in use in the system.
Delete Tag
Removes an unused tag from your system.
Download Document
Retrieves the original physical file for a selected document ID.
Get Correspondent
Fetches all details about an existing correspondent by name or ID.
Get Document
Retrieves the full metadata and OCR text for a specific document file.
Get Document Type
Fetches details about a defined document type category.
Get Tag
Retrieves the full definition of a tag by its ID.
List Correspondents
Shows a list of all correspondents currently tracked in your archive.
List Document Types
Displays all the document types and categories you have defined.
List Documents
Searches and lists documents across the entire archive, supporting filters by date...
Preview Document
Generates and shows a visual preview link for a selected document.
Thumb Document
Creates and retrieves a small thumbnail image for a document.
Update Correspondent
Modifies the details of an existing correspondent record.
Update Document
Changes the metadata (like tags or date) associated with a specific document.
Update Document Type
Modifies the definition of an existing document category.
Update Tag
Updates the description or name of a tag.
Upload Document
Ingests and processes a new physical file into your document archive.
Create Tag
Generates a new keyword or tag to classify documents.
Create Correspondent
Adds a new individual or company name used in your archive.
Create Document Type
Defines and adds a new category for documents, such as 'Invoice' or 'Lease...
Create Saved View
Saves a specific filter combination so you can easily recall that view later.
List Saved Views
Lists all predefined filter views that you have saved for quick access.
Delete Correspondent
Permanently removes an individual or company from your correspondent list.
Delete Document
Removes a specific document file from the archive.
Delete Document Type
Deletes an existing document category definition.
Delete Saved View
Removes a previously saved filter view.
Security and governance baked right in.
Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on each call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Paperless-ngx, then connect any of our 5,200+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,200+ others, all in one place
- Add new capabilities to your AI anytime you want
- Connections are secured and governed automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog weekly
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Paperless-ngx. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS CLOUD
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on each call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
The Problem: Document searches are always too slow.
Today, finding one piece of information means navigating multiple systems or running complex queries in a dashboard. You have to remember if the record is filed under 'Client' or 'Project,' and then manually cross-reference dates across different folders just to confirm who signed what.
With this MCP, your agent acts like an internal expert. You ask it directly: 'Who was responsible for the billing in March 2023?' The agent doesn't browse; it uses its tools—like `list_documents` and filtering by date—to pull the specific records and deliver the answer instantly.
Paperless-ngx: Full document lifecycle management with Paperless-ngx MCP
The tedious steps of manually creating tags, then having to log back in later to update a correspondent’s name on several records are gone. You simply ask the agent to `update_correspondent` or run `create_tag`, and it handles the API calls across your entire archive.
Your AI client now manages the entire document lifecycle for you, from initial upload via `upload_document` to final retrieval using `get_document`. It's a seamless transition from paper chaos to conversational data access.
What Paperless-ngx MCP does for your AI
Stop wading through endless folders just to find one date or name. This connector lets you treat your document archive like a librarian that knows every word in it. Instead of manually searching folder structures or running complex queries, you simply ask your agent for what you need—and get an answer based on the full text and metadata of your files.
You can use dedicated functions to upload new documents, organize them by creating tags or updating correspondents, or pull out a specific document's details. If you're managing records across different systems, connecting it through Vinkius gives you one place to access this power from any compatible client.
It’s about turning static files into actionable data.
019e38d2-4914-73e8-a4af-17ec62c81c35 How to set up Paperless-ngx MCP
The bottom line is you tell the AI what you want in plain English; it handles the complex API calls and retrieval logic for you.
Subscribe to this MCP and provide the necessary Paperless-ngx API URL and Personal API Token.
Connect your preferred AI client (Claude, Cursor, etc.) to Vinkius via the MCP connection point.
Directly ask your agent questions about your documents. The agent will call the appropriate tools to search, retrieve data, or execute actions.
Who uses Paperless-ngx MCP
Anyone dealing with large volumes of paper-based or digital records—legal teams, research academics, or small business operations managers. If your job involves finding specific clauses in old contracts or compiling reports from disparate sources, this is for you.
Retrieving all documents related to a single correspondent over a five-year period and verifying document types.
Uploading monthly utility bills, creating the correct tag for 'Utilities,' and updating the record's metadata automatically.
Listing all saved views or document types to ensure consistent classification across a historical collection of records.
Benefits of connecting Paperless-ngx MCP
Find exactly what you need without digging through folders. The list_documents tool allows your agent to search across keywords, dates, and tags instantly. No more manual filtering.
Consolidate all document data into one place. By using the MCP, your AI client can access every detail of a file via get_document, giving you full OCR text for analysis.
Maintain perfect order with metadata management. You can use tools like create_tag and update_document to ensure new files are classified correctly right away.
Visualize and manage your library instantly. The agent can provide a quick visual context using preview_document, so you don't have to download every file just to check it out.
Build robust reporting workflows. By listing all document types (list_document_types) or saved views (list_saved_views), your AI client builds reports based on consistent, predefined structures.
Paperless-ngx MCP use cases
Verifying Contract Dates
A paralegal needs to know the start date of every contract signed by 'Smith & Sons'. They ask their agent to list documents and filter by correspondent. The agent uses list_documents and get_correspondent to pull up all relevant files, providing a concise timeline instead of a folder dump.
Processing Bulk Invoices
An ops manager receives 50 invoices. Instead of manually uploading them, they use the agent to upload_document in batches. The agent then uses create_tag and update_document immediately after ingestion, ensuring every file is categorized as 'Invoice' and tagged with the correct vendor name.
Researching Historical Data
A researcher needs all documents related to 'Project Chimera' from 2018. They ask their agent for a list of documents, specifying both keywords and the date range. The agent calls list_documents, returning only the relevant records so they can start writing immediately.
Updating Client Records
A legal team updates client information. Instead of logging into two different systems, they tell their agent to update the correspondent details using update_correspondent, ensuring all archival files linked to that person reflect the current contact info.
Paperless-ngx MCP tradeoffs
What to watch out for, and the recommended way to handle each one.
Treating it like a file share
Asking your agent, 'Find me the folder for 2023 tax returns.' The agent can't navigate folders; it only speaks to structured data and metadata.
Instead, ask: 'List all documents related to taxes from 2023 using list_documents. Then use get_document on the ID you want details for.' Always reference criteria, not folder paths.
Manual Data Entry
Having to manually read a document and type out its key information (e.g., client name, date) into a spreadsheet.
Use the agent to call get_document on the file ID. The full OCR text is returned, allowing your AI client to extract and structure the data points for you automatically.
Ignoring existing categories
Treating every document as new and not giving it any context or type.
Before uploading, use create_document_type to define what a 'Warrant' is. Then, when you upload the file using upload_document, the system automatically categorizes it correctly.
When to use Paperless-ngx MCP
Use this MCP if your primary pain point is finding or organizing information locked inside structured documents—contracts, invoices, reports. If you need to know what happened based on a document's content (e.g., 'What was the monthly rent mentioned in this lease?'), this is ideal because of tools like get_document and OCR text retrieval.
Don't use it if your data lives outside Paperless-ngx, or if you are trying to summarize general knowledge that has nothing to do with a specific file archive. If your goal is simply conversational chat without document grounding, just use a standard LLM. This MCP adds the critical layer of 'document truth' to your conversations.
Frequently asked questions about Paperless-ngx MCP
How do I search my documents with Paperless-ngx MCP? +
You use the agent and ask it to list documents. The underlying tool, list_documents, allows you to specify filters like dates or keywords in your prompt.
Can I upload new files using the Paperless-ngx MCP? +
Yes. You use the upload_document tool. The agent handles sending the file and ensuring it gets processed by Paperless-ngx for OCR indexing.
What if a document needs more tags? Can I update them with Paperless-ngx MCP? +
You can use update_document to change metadata, or you can call the specific create_tag tool if the tag doesn't exist yet.
Do I need API tokens for Paperless-ngx MCP? +
Yes. You must provide your Personal API Token and API URL when subscribing to this MCP so the agent can authenticate and talk to your instance.
Can I retrieve just the text from a document using Paperless-ngx MCP? +
Absolutely. The get_document tool retrieves complete OCR text, allowing your AI client to read and summarize the content without needing a visual preview.
Can I search for documents using specific tags or date ranges? +
Yes. The list_documents tool allows you to filter by tags__id__in and created__date__gte. You can also perform a full-text search using the query parameter.
Is it possible to see a preview of a document without downloading the whole file? +
Absolutely. Use the preview_document or thumb_document tools to get visual representations of the document content directly through the agent.
Can I create new organization categories like tags or correspondents via AI? +
Yes, you have full management capabilities. You can use create_tag, create_correspondent, and create_document_type to organize your archive on the fly.