Automate invoice processing by uploading invoice images and extracting key invoice data, such as invoice number, type, language, items, prices, and total amount. Outputs results as a markdown table and structured CSV file for streamlined financial workflows.
User Uploads Invoice Image
User uploads an invoice image to the chatbot interface.AI Extracts Invoice Data
The workflow uses AI to perform OCR and extract invoice details, including number, type, language, items, prices, and total amount.Display Results in Chat
Extracted invoice data is presented to the user in a clear markdown table via chat.Generate Structured Data
The workflow further structures extracted data into fields like company name, VAT ID, service description, and amount.Export to CSV
Structured invoice data is exported as a CSV file for easy download and further processing.Below is a complete list of all prompts used in this flow to achieve its functionality. Prompts are the instructions given to the AI model to generate responses or perform actions. They guide the AI in understanding user intent and generating relevant outputs.
Prompt for extracting invoice data from image using OCR and outputting results as a table.
You are an expert assistant in doing OCR and extracting Information from an invoice in the image. Given the invoice, extract the following from the invoice
- Invoice Number
- Type of Invoice
- the Language of Invoice
- Items in Invoice
- Price per item in Invoice
- total amount of invoice
Make one row per invoice Item
OUTPUT THE RESULT ALWAYS IN A TABLE
Below is a complete list of all components used in this flow to achieve its functionality. Components are the building blocks of every AI Flow. They allow you to create complex interactions and automate tasks by connecting various functionalities. Each component serves a specific purpose, such as handling user input, processing data, or integrating with external services.
Flow description
This workflow, titled “Invoice data extractor OCR”, is designed to automate the process of extracting essential information from invoice images. It leverages OCR (Optical Character Recognition) and language models to present key invoice details in both a human-friendly markdown table and an exportable CSV format. This automation is particularly useful for businesses that handle large volumes of invoices, as it eliminates manual data entry, reduces errors, and increases operational efficiency.
When a user interacts with the chatbot interface, the following sequence of actions occurs:
Chatbot Initialization & Welcome
Invoice Image Upload
Data Extraction via Prompting
Once an image is uploaded, the workflow composes a structured prompt for an AI model. This prompt instructs the AI to extract the following details from the invoice:
The prompt specifically requests that the AI outputs the extracted information in a markdown table format, with one row per invoice item.
AI-Powered OCR and Information Extraction
Display and Export of Results
| Step | Component | Functionality |
|---|---|---|
| 1 | Chat Opened Trigger | Detects when a user opens the chat. |
| 2 | Message Widget | Sends a welcome and instruction message to the user. |
| 3 | Chat Input | Accepts user input, specifically the invoice image file. |
| 4 | Prompt Template | Prepares a detailed extraction prompt for the AI model, tailored for invoice data. |
| 5 | Generator | Runs OCR and data extraction on the invoice image based on the prepared prompt. |
| 6 | Chat Output | Presents the extracted information in a markdown table within the chat. |
| 7 | Structured Generator | Extracts additional structured data for CSV export (company name, VAT ID, service, amount). |
| 8 | CSV Output | Generates and provides a downloadable CSV file from the structured data. |
After processing an invoice image, the user receives a table similar to the following in the chat:
| Invoice Number | Type | Language | Item Description | Price per Item | Total Amount |
|---|---|---|---|---|---|
| 12345 | Sales | English | Widget A | $10 | $100 |
| 12345 | Sales | English | Widget B | $20 | $40 |
Additionally, the user can download a CSV file containing structured fields such as:
This workflow streamlines a historically manual and error-prone process. By automating the extraction and structuring of invoice data, businesses can save time, reduce costs, and focus their resources on higher-value activities. The dual outputs (chat table and CSV file) ensure both immediate visibility and long-term data usability, making the solution practical for teams of any size.
We help companies like yours to develop smart chatbots, MCP Servers, AI tools or other types of AI automation to replace human in repetitive tasks in your organization.
