Bodywork website and booking system
A website, Telegram bot, mini app, and API carry one visitor from understanding the specialist's work to booking without repeating the request.
- Visitor path
- A visitor understands the specialist's work on the public page, continues in Telegram, and reaches booking without starting the request again.
- Shared request state
- The website, bot, mini app, and API use the same data, so one inquiry stays a single request across every interface.
- What the client can recheck
- The client can identify what is live and recheck the complete path after a change or restart.
The project job
The public page explains the specialist's work, Telegram continues the same inquiry, and the booking step receives the context already supplied by the visitor.
What must work as one path
- A clear first contact: The website explains the specialist's work in concrete language and gives the visitor a clear next step.
- Booking without a reset: The bot and mini app continue the same client path, and the visitor does not have to reconstruct the request.
- One working version: The client can tell which site, bot, and API version is live and whether the complete path works.
- Repeatable recovery: The system can be started and checked again without reconstructing the procedure from old chat messages.
One path across the system
PUBLIC WEBSITE
dorodnyaya.ru → specialist context + next action
↓
TELEGRAM
bot + mini app → questions + booking path
↓
API + DATA
one request state across every interface
↓
LIVE SYSTEM
version identity | health checks | recoverable servicesWhat the complete path shows
- Public Site
- dorodnyaya.ru
- Entry Points
- site + bot + mini app
- Client Path
- first contact to booking
- Request State
- shared across interfaces
- Live Version
- identified and checked
- Support
- repeatable start and recovery
Why the parts stay together
- Problem
- A public page, Telegram bot, mini app, and API can easily drift into four unrelated products.
- Solution
- Keep them in one Rust codebase with shared data contracts and one repeatable way to start and verify the complete path.
- Why not the other option
- Building each channel separately is faster at first, but then the request state, service wording, and live version have to be reconciled by hand.
Why the public wording is shared
- Problem
- If the website and Telegram flow describe the specialist's work differently, booking starts by clarifying a request the visitor already made.
- Solution
- The website, bot, and booking path use the same service terms, non-medical scope, and next-step expectations.
Implementation
- Site
- Rust, Leptos, SSR
- Telegram
- Teloxide bot, Telegram Mini App
- API
- Axum, SQLx, PostgreSQL
- Operations
- Nginx, TLS, rootless Podman, systemd