What needed to be done.
The main task is to revisit assets marked during analysis and track long-term investments. Instead of checking several accounts, the system brings balances, history and important changes into one view.
- Connected data collection from exchanges, wallets and the broker account.
- Separated data retrieval, checks and display in the app.
- Built portfolio history and assets by account, with funding balances tracked separately.
- Added warnings for errors and missing prices, access controls and AI-assisted news analysis.
How it works.
- Collect account state
The system retrieves information from accounts and wallets. Source responses and errors are saved to help investigate failures.
- Prepare one portfolio view
The system combines account data and saves portfolio history.
- Review assets and events
The user sees allocations, data gaps, alerts and analyzed news.
Missing data is visible.
If required data is unavailable, the system does not save a portfolio total. Errors in optional sources are shown as warnings.
News analysis is checked.
AI prepares an event summary. The response is checked before notification, and frequency limits help avoid repeated alerts.
Implementation and example.
The original UI from the safe public version uses fictional assets, history and events. Real accounts and balances are not shown.
The demo lets you browse portfolio tabs. It does not run collectors, the analyzer or trading operations.
DIAGRAM OF THE IMPLEMENTED SYSTEM
- Exchanges, wallets and broker
Collectors account for source-specific formats and behavior.
- A combined account overview
State and history preparation with required-read checks and error handling.
- Portfolio and market events
Assets by account, history, alerts and news summaries to review.
Architecture diagram of the project. Live data and connections are not shown.
The result.
Accounts and assets are available in one app with history and events. Missing asset prices are shown as gaps in the data. The person using the system makes portfolio decisions.
Technologies and implementation details
n8n, JavaScript, exchange and wallet APIs, Supabase/PostgreSQL, Hermes and a Telegram Mini App. Collectors, staging, normalization and the read layer are separate. The Mini App shows state; the AI pipeline prepares structured event analysis.
Let's discuss your task.
If your data is scattered across accounts and spreadsheets, show me the overview you need. We can discuss sources, indicators and events worth flagging.
Discuss a monitoring system ↗︎