Affinda MCP. Extract structured data from resumes, invoices, and IDs.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Affinda. This tool lets your AI agent parse structured data from documents like resumes, invoices, and passports. Upload PDFs or images and get high-accuracy JSON output.
You can also audit document statuses across specific workspaces and manage your parsing models through natural conversation. It handles structured data extraction and document management.
What your AI agents can do
Create document
Uploads and parses a PDF or image into Affinda to extract structured JSON data.
Get document
Retrieves the full structured JSON data and current status for a specific document ID.
List document types
Lists the specific document types (e.g., Resume, Invoice) your Affinda account supports.
Upload a PDF or image, and the tool extracts structured data into a high-accuracy JSON format.
Retrieve the structured JSON data and processing status for a specific document ID.
Retrieves a list of document types (like Resume, Invoice, or Passport) that Affinda supports.
Retrieves a list of all documents within a specified workspace, including their processing status.
Retrieves all the container workspaces that hold documents in your Affinda account.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
Affinda MCP Server: 5 Tools for Document Parsing
These tools let your agent manage document lifecycles, from listing workspaces to parsing complex files into structured JSON.
019d7548create document
Uploads and parses a PDF or image into Affinda to extract structured JSON data.
019d7548get document
Retrieves the full structured JSON data and current status for a specific document ID.
019d7548list document types
Lists the specific document types (e.g., Resume, Invoice) your Affinda account supports.
019d7548list documents
Retrieves a list of all documents in a workspace along with their processing status.
019d7548list workspaces
Retrieves all the container workspaces where documents are stored in your Affinda account.
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 every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Affinda, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
What you can do with this MCP connector
Affinda MCP Server: Structured Data Extraction
Connect your AI agent to this server, and it'll let you pull structured data from documents like resumes, invoices, and passports. You upload a PDF or an image, and the tool spits out high-accuracy JSON. You can also check document statuses and manage your parsing models just by talking to your agent.
create_document: Uploads and parses a PDF or image into Affinda, giving you structured JSON data. get_document: Retrieves the full structured JSON data and current status for a specific document ID. list_document_types: Lists the specific document types your Affinda account supports, like Resume or Invoice. list_documents: Retrieves a list of all documents in a workspace, along with their processing status. list_workspaces: Gets all the container workspaces where your documents live in Affinda.
How Affinda MCP Works
- 1 Subscribe to the server and provide your Affinda Bearer Token.
- 2 Your AI agent sends a command to parse a document (PDF or image) via its public URL.
- 3 The server processes the document and returns the resulting structured JSON data and status.
The bottom line is, you tell your agent what document to parse, and it handles the connection, the extraction, and the structured data output.
Who Is Affinda MCP For?
This is for operations managers, compliance officers, and HR/finance teams. If your job involves reviewing documents—whether it's checking a candidate's resume or verifying an invoice total—you need this. It takes the manual, error-prone work of data entry and turns it into a conversational command.
Runs resume screening and extracts candidate contact info from batches of CVs. Needs to know the metadata extracted from the documents.
Audits invoice data extraction, verifying payment details and ensuring the right fields were pulled before submitting payment.
Monitors document processing workflows, listing all documents across different workspaces to ensure nothing is stuck or unprocessed.
Retrieves identity documents (like passports or IDs) to perform risk assessments and verify data points for regulatory compliance.
What Changes When You Connect
- Stop manual data entry. Upload a batch of resumes and use
create_documentto get structured JSON data instantly. Your agent handles the parsing, giving you metadata for every candidate. - Audit document flow. Use
list_documentsto see every file in a workspace and check its processing status. You never have to manually check if a document is stuck or incomplete again. - Verify data sources. Run
list_document_typesto know exactly which models (Invoice, Receipt, Passport) your account supports. This prevents you from trying to parse unsupported files. - Track documents across teams. Use
list_workspacesto see all the document containers in your account. This gives a clear map of where every type of data lives. - Get instant data deep dives. After parsing, use
get_documentto pull the complete structured JSON for a specific file. You get the data, not just a status message. - Focus on insight, not process. By letting your agent manage the parsing and status checks, you spend time analyzing candidate data instead of managing document lifecycles.
Real-World Use Cases
Onboarding a new hire
The HR manager needs to process a new candidate's CV and pay stubs. They ask their agent to 'Parse this resume and the attached W-2.' The agent uses create_document with both files, and the server returns clean JSON data for the candidate's name, years of experience, and salary details. The manager gets the structured data immediately.
Reconciling vendor invoices
The Accounts Payable specialist has 50 invoices to check. Instead of opening 50 PDFs, they tell their agent to 'List all documents in the AP workspace.' The agent uses list_documents to verify the status, then uses get_document on the ones marked 'Ready' to pull the invoice totals and vendor details for bulk verification.
Compliance audit of IDs
A compliance officer needs to check if all employee files contain a valid passport. They ask the agent to list all available document types, confirm 'Passport' is supported, and then run create_document on the required IDs. The agent returns structured JSON containing the key identity fields for risk assessment.
Auditing document storage
The operations manager needs to know if the 'Q3 Financial' workspace has any documents. They ask the agent to 'List all workspaces' and then 'List documents in Q3 Financial.' This confirms the scope of data and ensures no critical documents were accidentally left outside a monitored folder.
The Tradeoffs
Trying to parse everything at once
The developer tries to send one giant request to parse a mix of invoices, resumes, and passports simultaneously, hoping the server handles it all. This fails because the system needs a specific model for each document type.
→
First, use list_document_types to confirm the available models. Then, send targeted requests: use create_document for the resume model, and a separate create_document for the invoice model. Keep the calls specific.
Assuming status means 'ready'
A user sees a document in their list and assumes it's fully parsed and ready to use. They try to access the data immediately, but it's actually still processing.
→
Always check the status first. Use list_documents to get the document ID, then use get_document to verify the processing status before attempting to use the data.
Manually listing every document ID
The user manually tries to retrieve the data for 100 documents one by one by calling get_document 100 times. This is slow and tedious.
→
Use list_documents to get a manageable list of document IDs and their statuses. Then, if needed, your agent can iterate over the list and run get_document in batches or use the list results to filter for ready files.
When It Fits, When It Doesn't
Use this server if your core task is taking unstructured documents (PDFs, images) and turning them into structured, queryable JSON data. It's essential for processes like screening candidates, verifying financial records, or auditing identity files. Don't use it if your goal is to manage the physical storage of documents, or if you need real-time collaboration features (like co-editing). For simple document storage, use a cloud file system API. If you only need to know where the document is, but not what is inside, use a basic file system listing tool instead.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Affinda. 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 INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 5 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Data extraction shouldn't require specialized training or multiple copy/paste sessions.
Today, if you need to extract data from a mix of documents—say, a resume and an invoice—you open the PDF, manually find the name, copy it, switch to the invoice, copy the invoice number, and then paste both into a spreadsheet. This process takes minutes, and you're guaranteed to miss something or copy the wrong field.
With the Affinda MCP Server, you just tell your agent to 'Parse these documents.' It handles the entire flow: it uploads the files, uses the correct model for the resume and the invoice, and returns clean, structured JSON data that you can read directly into your workflow.
Affinda MCP Server: Get structured data from documents
The manual steps that disappear include: opening the document, locating the field (e.g., 'Date of Birth'), deciding if it's correct, and manually typing it into a record. You don't click through a dozen tabs or write complex parsing scripts.
Now, your agent talks to the server, and the server does the heavy lifting. You get the final, usable data set without touching a spreadsheet or a single PDF.
Common Questions About Affinda MCP
How do I use the Affinda MCP Server to parse a resume? +
You use the create_document tool. You provide the URL to the resume PDF or image, and the server uses the 'Resume' model to extract structured JSON data for fields like name and experience.
What is the difference between `list_documents` and `list_workspaces` in Affinda? +
list_workspaces shows the top-level containers where your documents live. list_documents lists the actual files inside one of those containers, along with their current processing status.
Can the Affinda MCP Server handle multiple document types? +
Yes. You can run create_document on multiple files in one session, provided you use the correct model for each file (e.g., one model for 'Invoice' and another for 'Passport').
How do I check if a document parsing job is finished? +
You use get_document. This tool retrieves the status and the fully structured JSON data for a specific document ID. The status field tells you if the process is complete.
How do I use the `list_document_types` tool in Affinda to see what models I can use? +
The list_document_types tool shows exactly which parsing models your Affinda account supports. This lets you confirm if you can process documents like receipts, passports, or resumes.
What does the `get_document` tool do if a document fails to parse in Affinda? +
The get_document tool provides the full structured JSON data and the status of the document. If parsing fails, you'll see the technical metadata and the specific error message.
If I need to check my document library, should I use `list_documents` or `list_workspaces`? +
list_documents retrieves all parsed files within a specific workspace. Use this when you want to see the status and metadata of the actual documents you uploaded.
How do I start a new document parsing job using the `create_document` tool? +
You use the create_document tool by uploading a PDF or image via its public URL. This initiates high-accuracy JSON extraction from the document.
How do I find my Affinda API Key? +
Log in to Affinda, click on your profile avatar, and go to Settings > API Keys. You can generate a new token there. Remember to copy it immediately, as it is only shown once.
What document types are supported? +
Affinda supports dozens of specialized models including Resumes, Invoices, Receipts, Passports, Driver Licenses, and Purchase Orders. Use the list_document_types tool to see the full list for your account.
Can I parse a document via a public URL? +
Yes! The create_document tool accepts a url parameter. Simply provide the direct link to the PDF or image, and Affinda will retrieve and parse it automatically.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
YouCanBook.me
Automate scheduling workflows via YouCanBook.me — manage booking pages, appointments, and availability directly from any AI agent.
Referly
Automate affiliate programs via Referly — manage partners, track conversions, and monitor workflows directly.
Huawei Push Kit / 华为推送服务
Huawei's essential device push platform — send notifications to HMS devices via AI.
You might also like
Gripr
Manage safety compliance, inspections, and risk assessments for construction and industrial projects with mobile-first tools.
JobProgress (Leap)
Manage contractors, customers, and jobs via JobProgress API.
Dot Object Transformer
Flatten deeply nested JSON objects into dot-notation keys, or reconstruct nested objects from flat dictionaries.