← Вернуться к списку постов

Automating Counterparty Verification: From a Contract Scan to a Ready Document

Статья | 22 January 2026

Manually entering details from a contract into a CRM or document templates is not only tedious but also prone to errors. It's even more challenging to quickly verify a counterparty's status: whether it is active, in the process of liquidation, or bankrupt. However, you can make this process fast, accurate, and free from constant oversight.
You can build such a solution by combining the capabilities of the low-code platform Botman.one with modern neural networks and the API of the Dadata service. You can create a service that automatically extracts details even from a scanned contract, verifies the counterparty via the Dadata service, and inserts the correct data into any documents.

How It Works: The Automation Workflow

  1. Document Upload. The user uploads a contract (PDF or even a scanned image) via a form in a Telegram bot, on a website, or in the Botman.one interface.

  2. Text Extraction (OCR). If the document is an image, the system uses an OCR service, built into a neural network like DeepSeek, to recognize the text. This step is skipped for text files.

  3. Extracting Details. The neural network analyzes the text and finds key data: Tax ID (INN), Primary State Registration Number (OGRN), company name, address, bank details.

  4. Counterparty Verification. The extracted Tax ID (INN) is sent to the Dadata service API to obtain current data from the Unified State Register of Legal Entities (EGRUL) and check the company's status (active, in liquidation, bankrupt).

  5. Document Generation. The verified details are automatically inserted into pre-prepared document templates in the Botman.one constructor, for example, into an act, a new contract, or an invoice.

  6. Result. The finalized document and the verification results are sent to the user.

Technical Implementation: Key Components

1. Working with Documents and Neural Networks in Botman.one
The Botman.one platform is ideal for orchestrating the entire process. You can use the built-in functionality for file uploads (the "additional info" block), and for interacting with the neural network API and the Dadata service API, use the "Action" block. It allows sending HTTP requests to any external API.
For Russian users, it's most convenient to send requests to GigaChat (after registering and obtaining a key there). If you prefer to use foreign neural networks, you can use the Proxy API service to pay for access in rubles, obtain a key from Proxy API, and insert it into the corresponding field in the "Action" block settings.

  • Text Extraction (OCR): In Botman.one, you configure a request to a neural network API capable of recognizing text in an image (e.g., DeepSeek), pass the file, and receive the recognized text. For this, in the "Action" block, you need to select the "AI request" type and configure the parameters: insert the API key for the neural network obtained from the Proxy API service (for foreign models) or a GigaChat key.

  • Text Analysis by AI: With a properly composed prompt, you can accurately extract structured data from the document. For example: "From the following contract text, extract the INN, OGRN, full legal name, legal address, and bank account. Output the result in JSON format."
    This text needs to be inserted into a pre-created "Message" block. Then, after creating the "Action" block for sending the document to the neural network, you need to configure it to:

    • Take the document file for analysis from the "Additional Info" block (your service user will upload the file there).

    • Take the prompt for the neural network from the "Message" block.

2. Verification via Dadata API
Dadata provides an API for obtaining data about organizations by INN or OGRN. The response contains not only current details but also the key company status: ACTIVE, LIQUIDATING, LIQUIDATED, REORGANIZING, BANKRUPT.
In Botman.one, you configure an external request to the Dadata API also via the "Action" block, inserting the INN extracted by the neural network into the settings. The response is parsed, and the company status is saved into a variable for further use.

3. Building the Complete Solution
All these stages are linked into a single scenario (funnel) within Botman.one. The platform allows you to:

  • Handle logic branching (e.g., if the company status is not "ACTIVE", send a warning).

  • Automatically fill document templates with verified data.

  • Send the final files to the user via email, Telegram, or save them to cloud storage.

Advantages of This Approach

  • Speed. A process that took hours is completed in minutes without human intervention.

  • Accuracy. Eliminates typos when transferring details. Data is always up-to-date and sourced from the official registry (EGRUL).

  • Security. Automatic company status verification minimizes the risks of working with unreliable counterparties.

  • Flexibility. The scenario can be adapted to any document type and business process: from counterparty verification in procurement to automatic contract generation in sales.

  • Accessibility. Creating such a solution does not require programming knowledge – just an understanding of the process logic and the ability to work with Botman.one's low-code tools.

Conclusion
Integrating Botman.one with neural networks and external APIs opens up powerful opportunities for automating routine and critical tasks. The described counterparty verification case is just one example. The low-code platform allows for rapid prototyping, testing, and implementation of such solutions, significantly increasing efficiency and reducing operational risks.

Ready to automate your document workflow?
Start by exploring the capabilities of Botman.one and creating your first bot.