# Alphamoon MCP MCP > Alphamoon extracts structured data from contracts, invoices, and forms using AI-powered OCR. This MCP lets your agent automate document processing workflows by pulling specific fields—like total amounts or vendor names—from complex documents without manual entry. ## Overview - **Category:** productivity - **Price:** Free - **Tags:** alphamoon, ocr-api, document-extraction, intelligent-data-processing, invoice-automation, nlp-tools, data-orchestration, ai-automation, mcp ## Description Handling a stack of scanned receipts or PDFs is usually a nightmare; you're staring at unstructured data that takes hours to transcribe. With this connection, your AI client acts like a dedicated document engineer. It doesn't just read the file; it understands what kind of data it is—a date, an invoice total, a specific clause—and pulls it out as clean fields. You can manage entire pipelines, list available templates, and monitor every step of the process without leaving your chat window. If you're used to copy-pasting numbers from PDFs into spreadsheets, this changes that. Your agent handles the full document lifecycle, giving you high-fidelity data and confidence scores for everything it finds. This capability is hosted on Vinkius, which makes connecting Alphamoon with any other tool in its catalog simple. ## Tools ### check_alphamoon_status Checks the connection status of your Alphamoon account for immediate operational verification. ### delete_document Removes a specified document record from the system's storage. ### get_document Retrieves specific metadata and details about an existing document file. ### get_extraction_results Fetches the structured, extracted data points from a processed document based on its ID. ### get_ocr_results Retrieves the complete, high-fidelity text recognized from scanned documents or images. ### get_process Gets detailed information about a specific document automation pipeline configuration. ### get_template Retrieves the details of an available extraction template, showing what data fields it expects to find. ### list_documents_by_process Filters and lists documents that were processed using a specific automation pipeline ID. ### list_documents Lists all documents currently stored in your Alphamoon account. ### list_processes Lists all defined document processing pipelines available for use in your workflows. ### list_templates Lists all custom data extraction templates you have created or that are available. ### upload_document Uploads a new document file into the system for processing and analysis. ## Prompt Examples **Prompt:** ``` List all my document automation pipelines in Alphamoon. ``` **Response:** ``` I've retrieved your automation directory. You currently have 3 active pipelines: 'Invoice Processing' (ID: pipe_123), 'ID Verification', and 'Receipt OCR'. Which one should we check for processed documents? ``` **Prompt:** ``` Show the extraction results for document ID '90210'. ``` **Response:** ``` Fetching results... For document 90210, I've identified high-fidelity fields including 'Total Amount' ($250.00), 'Date' (2026-04-24), and 'Vendor Name'. All fields have 99% confidence scores. Need the raw OCR text? ``` **Prompt:** ``` Get the raw OCR text from document '90210'. ``` **Response:** ``` Accessing OCR data... I've retrieved the complete high-fidelity recognized text from document 90210. It contains 15 paragraphs and includes all detected hand-written notes. Shall I summarize the content for you? ``` ## Capabilities ### Extract Structured Fields The agent identifies specific pieces of information—like a total amount or an ID number—and returns them as organized data points, rather than just raw text. ### Manage Document Pipelines You programmatically list and coordinate entire document automation workflows to ensure high-fidelity extraction across different types of files. ### Access Raw OCR Text The agent retrieves the complete, recognized text from scanned documents, including any handwritten notes or non-standard formatting. ### Process and Track Documents You can list available templates; process pipelines; and monitor individual document statuses to maintain a perfect audit trail for every file processed. ### Handle Document Lifecycle The agent allows you to upload documents, check the system status, or delete records when they are no longer needed. ## Use Cases ### Processing a mixed batch of receipts An ops manager uploads 20 random receipts. Instead of manually opening each one, the agent calls `list_processes` to find the receipt template, then uses `get_process` to confirm its settings, finally calling `get_extraction_results` on all 20 documents simultaneously. ### Auditing historical invoices A financial analyst needs data from last quarter. They use `list_documents` to pull a list of document IDs; then, they iterate through those IDs using `get_document` to confirm the metadata before running extraction on the most critical ones. ### Building an automated ingestion service A developer needs reliable input for their custom application. They use `list_templates` to find the right data structure, then they call `get_template` details and finally use `upload_document` to feed it into a controlled pipeline. ### Data cleansing after migration After an old system dump, there are junk files. An administrator uses `list_documents` to find the IDs of obsolete records and then calls `delete_document` on those specific file IDs for proper data cleanup. ## Benefits - Stop manually reviewing logs; use `get_extraction_results` to instantly see structured fields and confidence scores for a processed document, verifying accuracy in natural language. - You can coordinate entire workflows by listing available automation pipelines using `list_processes`, ensuring the right process runs on the correct document type every time. - Need to check if your connection is working? The `check_alphamoon_status` tool gives instant operational reporting, telling you whether the API key and service are active before you start a big batch job. - Instead of searching through directories, use `list_documents_by_process` to filter and find only those documents that ran through your specific 'Invoice Processing' pipeline. - Don't just trust the summary; retrieve the full context by calling `get_ocr_results`. This gives you all the raw text from scanned files, including any handwritten notes for manual audit checks. ## How It Works The bottom line is that your AI agent turns vague files into precise, usable data structures within a single conversation. 1. First, subscribe to this MCP and retrieve your API Key from the Alphamoon dashboard. 2. Next, instruct your AI client to start processing a document; it will coordinate data extraction using available pipelines or templates. 3. Finally, you receive structured JSON output with identified fields—like 'Invoice ID' and 'Total Amount'—and confidence scores for each piece of data. ## Frequently Asked Questions **How do I start processing a new invoice with Alphamoon MCP?** You first upload the document using `upload_document`. After that, you need to select or confirm which data template applies by checking the available options via `list_templates`. **Can I retrieve handwritten notes from a PDF using get_ocr_results?** Yes. The `get_ocr_results` tool pulls all high-fidelity recognized text, including detected handwriting and non-standard print, giving you the full raw context. **What if I want to process only a specific type of file? Is there an Alphamoon MCP tool for that?** Yes. You can use `list_documents_by_process` to filter and list documents based on the automation pipeline they ran through, helping you target your analysis. **Is it possible to check if my API key is working before I start a job? (check_alphamoon_status)** Absolutely. Use `check_alphamoon_status` first. It verifies the live connectivity and operational status of your account, saving you time by catching issues upfront. **I need to clean up old files; how does the `delete_document` command work?** You use `delete_document` by providing the document ID you want removed. This tool permanently deletes the file and all associated metadata from your account, helping you maintain a clear audit trail. **Before running an extraction, how can I check the specific structure using `get_template`?** Call `get_template` to view the schema definition for any template. This lets you confirm exactly what fields (like 'Invoice Date' or 'Total Amount') your pipeline expects before you upload a single document. **I'm setting up a new workflow; how can I check all my active document automation pipelines with `list_processes`?** `list_processes` shows every configured automation pipeline ID. This is your starting point for understanding the full scope of workflows available to process different types of documents. **What information can `get_document` provide about a document's current processing status?** `get_document` returns detailed metadata, including the file's current status (e.g., 'Processing', 'Completed', or 'Failed'). It’s essential for monitoring long-running jobs and troubleshooting errors. **How do I find my Alphamoon API Key?** Log in to your account, navigate to **Settings** > **API**, and generate a new API Key for your integration. **Can I retrieve raw OCR text via AI?** Yes! The `get_ocr_results` tool allows your agent to retrieve the complete high-fidelity recognized text from any processed document. **How do I check extraction results?** Use the `get_extraction_results` tool and provide the document ID to retrieve structured fields and confidence scores programmatically.