Installation

Trading Journal AI runs locally on your computer. Setup is a few commands in a terminal — you don't need to be a developer, but you will copy and paste a little.

What you'll need

Node.js 20 or newer and npm (they come together).
Git and a terminal to run commands in.
Optional: a free Massive API key for candlestick chart data. You can skip it and add it later — the app still runs, you just won't be able to fetch new chart candles until you add one.

Install

Open a terminal and run:

git clone https://github.com/trading-journal-ai/trading-journal.git
cd trading-journal
./install-trading-journal.sh

Run the installer from inside the trading-journal folder. If you see no such file or directory, you're probably one folder too high — run cd trading-journal first, then try again.

The installer walks you through everything and, when it's done, starts the app and prints the local address (usually http://localhost:3000). Along the way it:

installs the project's dependencies,
lets you choose demo data or an empty journal for your own broker CSV,
lets you add a Massive API key or skip chart fetching for now,
creates the local SQLite database, and
starts the app on localhost.

Nothing is installed globally — everything it creates stays inside the project folder.

Two ways to start

The installer offers two modes:

Trading Journal Demo — loads sample trades and notes so you can preview the app immediately.
Trading Journal — creates an empty local journal, ready for your own imported trades.

Where your data lives

Everything you create is stored on your machine and kept out of version control. The gitignored files include:

data/journal.db        # your journal (SQLite)
data/uploads/          # attachments
.env.local             # your Massive API key

For a backup, copy the data/ folder somewhere safe. Real broker exports should never be committed — they can contain account numbers and full trade history.

Next, walk through your first review in the Quick Start.