# AlgoDocs MCP for AI Agents AI Agent Connect > AlgoDocs MCP lets your AI agent handle heavy-duty document extraction. It pulls data from PDFs, images, and Word files into structured JSON. You can manage your extraction rules, organize folders, and monitor your usage stats without leaving your chat interface. It's built for teams that need to turn piles of documents into usable data quickly. ## Overview - **Category:** artificial-intelligence - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_0QGDA1pBXpGO1Q5IE0Ud8GMVJIzU6NBeEo6fN1bS/ai-agent-connect - **Tags:** ocr, data-extraction, pdf-parsing, automation, invoice-processing, machine-learning ## Description When you connect AlgoDocs to your AI client, you stop copy-pasting data from messy PDFs or scanned receipts into spreadsheets. Instead, you just give your agent a URL or a file, and it handles the heavy lifting of turning that content into a clean JSON object. You can tell your agent to check the status of a batch of documents or ask it to find specific data points from a folder of invoices. If you need to see how your extraction rules are performing, your agent can pull up the details for your specific extractors or show you your total usage stats. It makes the whole process of moving data from a document into a database feel like a conversation rather than a manual data entry job. You can organize your projects into folders and keep everything tidy as you scale. Getting this set up is easy because Vinkius makes it simple to connect your account and start running these workflows immediately. It's a way to make your document pipeline actually work for you instead of making you work for it. Whether you're dealing with thousands of shipping labels, complex financial statements, or stacks of HR resumes, the goal is the same: getting the data out of the file and into your system without the headache of manual entry. You can manage your entire folder hierarchy, monitor your processing stats, and audit your models to ensure everything is running exactly as planned. This setup turns your AI agent into a reliable data processing worker that understands your specific extraction rules. ## Tools ### get_my_account Check your current account status and details. This helps you stay on top of your subscription. ### get_document_data Pull the final structured JSON results for a specific document. This is how you get the actual data out. ### list_extractor_data Get a bulk list of results for all documents processed by an extractor. This is perfect for checking large batches. ### get_document_status Find out if a document is still being processed or is ready for use. Use this to track your active jobs. ### get_api_usage Check your current usage statistics and account limits. This keeps your costs and limits visible at all times. ### get_folder_details View metadata and details for a specific storage folder. This helps you see the specifics of a project folder. ### list_recent_documents See a list of the most recently parsed documents in your account. This gives you a quick look at your latest work. ### list_extractors See a list of all your active AI extractors and their rules. Use this to choose the right model for a job. ### list_folders View the hierarchy of your document storage folders. This keeps your projects organized and easy to find. ### upload_document_from_url Send a URL to your agent to start parsing a document into JSON. It handles the extraction automatically. ## Prompt Examples **Prompt:** ``` List all the extractors I have set up. ``` **Response:** ``` I've pulled up your active extractors. You have 4 models ready: * **Invoice_Parser_V2** * **Receipt_Scanner** * **HR_Resumes** * **Shipping_Labels** Which one should we use for this batch? ``` **Prompt:** ``` Check the status of my latest batch. ``` **Response:** ``` I'm checking the status for your recent documents. It looks like **45 out of 50** have finished processing. The remaining 5 are still being parsed. Would you like me to grab the data for the finished ones? ``` **Prompt:** ``` Summarize my recent usage. ``` **Response:** ``` Here is your current usage summary: | Metric | Value | | :--- | :--- | | **Total Extractions** | 1,240 | | **API Usage** | 85% of monthly limit | | **Status** | Active | Your last 12 documents were processed in the last hour. ``` ## Capabilities ### Extract structured JSON from documents Parse PDFs, images, and Word docs into clean, usable data. ### Manage document folders Organize your extraction projects into a logical folder hierarchy. ### Audit extraction models Check the rules and details of your specific AI extractors. ### Monitor account usage See your API stats and account status in real time. ### Track processing status Get updates on whether your documents are finished or still being parsed. ## Use Cases ### Automated Invoice Processing A finance clerk gives the agent a folder of 500 invoice URLs. The agent confirms they are processed and ready for the accounting system. ### Resume Parsing at Scale An HR specialist asks the agent to parse a batch of resumes. The agent turns them into a structured list of candidates. ### Receipt Auditing An operations manager asks the agent to check the status of recent receipts. The agent uses list_recent_documents to provide a summary. ### Data Payload Retrieval A developer asks the agent to pull the data for a specific file. The agent uses get_document_data to show the specific JSON payload. ### Folder Metadata Inspection A manager wants to see the metadata for a specific project folder. The agent uses get_folder_details to show the folder's specific attributes. ## Benefits - Stop manual data entry by using upload_document_from_url to turn PDFs into JSON automatically. - Organize your entire project history by managing your hierarchy with list_folders. - Audit your extraction rules quickly using list_extractors to ensure your models are accurate. - Monitor your costs and limits in real time with get_api_usage and get_my_account. - Get bulk data for your entire pipeline at once using list_extractor_data instead of checking files one by one. - Track your progress instantly with get_document_status so you know exactly when your data is ready. ## How It Works The bottom line is you turn your AI agent into a high-speed document processing engine. 1. Subscribe to this Connector and enter your AlgoDocs email and API key in your client settings. 2. Provide your agent with a document URL or a Base64 string to start the extraction. 3. Receive a structured JSON payload or a status update directly in your chat. ## Frequently Asked Questions **Can AlgoDocs MCP handle scanned images?** Yes, it handles scanned images and PDFs to turn them into structured JSON data. **How does AlgoDocs MCP handle complex tables?** It's designed to parse complex tables within documents so you don't have to manually transcribe them. **Can I use AlgoDocs MCP for my invoices?** It's a popular choice for automating invoice and receipt data extraction into your existing systems. **Does AlgoDocs MCP support folder organization?** Yes, you can list and audit your folder hierarchy to keep your different projects organized. **Can I check my API limits with AlgoDocs MCP?** You can ask your agent to check your usage stats and account details directly through the chat. **Does AlgoDocs MCP work with Word docs?** Yes, it can parse Word documents just like it handles PDFs and images. **How do I find my AlgoDocs API Key?** Log in to AlgoDocs, go to your **Account Settings**, and you will find your API key there. You will need this along with your registered email for authentication. **What is an 'Extractor' in AlgoDocs?** An Extractor is a set of rules and AI models configured to pull specific fields from a certain type of document (e.g., an Invoice Extractor). You must specify an `extractor_id` when uploading documents. **Can I retrieve data from a previously processed document?** Yes! Use the `get_document_data` tool and provide the unique `document_id`. Your agent will retrieve the extracted structured data from AlgoDocs storage.