A personal analyst for managing one’s own portfolio. It exists so that investment decisions rest on current data and can be traced back later together with the context they came from.
What it does
- Broker data: portfolio, watchlist, earnings and dividend calendars.
- Regular reports: a morning overview on weekdays, a weekly calendar and Friday exports – by e-mail or via Telegram.
- An analytical framework in three steps: quantitative scoring of the watchlist, a pause to decide on priorities, and in-depth analysis of selected stocks.
- AI assistant integration via an MCP server, so the data can be worked with in a conversation.
The system never trades on its own – every decision stays with a human.
Where it is heading
A stable workflow for a single investor; next, the decision history available over MCP and remote access for a web-based AI assistant. The architecture should not stand in the way of later extending it to multiple accounts.