RU
All work

Мониторинг закупок · Telegram-бот

AGGA Tenders

A Telegram service for searching and continuously monitoring real EIS procurement notices under 44-FZ and 223-FZ, with filtering, contract history and duplicate protection.

My roleArchitecture, Telegram bot, data sources, testing and infrastructure diagnostics
FormatProcurement monitoring service prototype
AGGA Tenders — project screen 1
AGGA Tenders — project screen 2

Deliverables

New notices from the official registry become structured Telegram alerts, without duplicates or repeated manual searches.

  • Live EIS procurement search under 44-FZ and 223-FZ
  • Filtering by keyword, customer and tax ID
  • Notice number, subject, value, date, status and official source link
  • Contract and agreement history for 2025
  • Clear separation of initial and final contract values
  • Scheduled background checks and automatic alerts
  • First-run baseline and duplicate-notification protection
  • SQLite persistence across service restarts
  • Keyword and notification administration inside Telegram
  • Secure administrator binding with a one-time code
  • Pagination, network-error handling and health checks
  • Modular source layer with a safe Avtoban marketplace placeholder

Challenge

A supplier of road and construction machinery parts needs to monitor EIS across many search phrases, specific customers and tax IDs. Manual checks are time-consuming, new notices are easy to miss, and the initial maximum price does not necessarily match the final contract value. The automation also has to use authorised sources only and protect administrative settings.

Solution

I developed a modular Telegram bot with Python and aiogram 3. It searches procurement notices by keywords and tax ID, separates 44-FZ and 223-FZ data, links to the official record and provides 2025 contract history. Background monitoring creates a baseline on first launch, alerts only on new matches and recovers from network failures. SQLite stores notices, settings, administrator identity and notification history, while a one-time code secures the initial administrator binding.

Outcome

The prototype was verified against live EIS data, including Wirtgen and VOLVO searches, a Moscow public customer by tax ID, and 44-FZ/223-FZ contract history. Automated tests cover filtering, persistence, deduplication, alerts, administrator binding and error handling. systemd and Docker deployment configurations are prepared; infrastructure diagnostics identified selective Timeweb blocking and confirmed EIS access from Selectel. A safe adapter for the Avtoban marketplace awaits an official API or data feed, with no CAPTCHA or access-control bypasses.

Tools and integrations

Pythonaiogram 3SQLitepytestЕИС «Госзакупки»44-ФЗ223-ФЗsystemdDocker

Have a similar project?

Discuss a project