Quickstart

This is the five-minute path from a fresh install to a watchlist that pings you when conviction shifts.

Install

Cents isn’t on PyPI yet — install from source:

git clone https://github.com/wolfbane/cents.git
cd cents
pip install .                  # Basic install
pip install ".[dev]"           # With test dependencies
pip install ".[broker]"        # With Alpaca trading integration

Cents requires Python 3.11+. You’ll also want at least an FMP API key — see Configuration for the full key matrix.

The five-step loop

1. Create a thesis. A thesis is a written hypothesis: what you believe, how you’d value it, and the time horizon you’re playing for. Everything else hangs off this record.

   cents thesis create --symbol NVDA --title "NVDA AI dominance continues" \
     --hypothesis "Data center growth will drive earnings beats" \
     --valuation undervalued --time-horizon medium

2. Add the symbol to your watchlist. Watchlist entries link a symbol to a thesis and an alert threshold (in conviction-delta points). The scan command will fire when crossed.

   cents watch add NVDA --thesis <thesis-id> --threshold 5.0

3. Run research. This is where the agents earn their keep — fundamentals, technical, macro, sentiment, moat, and insider all gather evidence and the orchestrator weights them into a single conviction delta.

   cents research NVDA --thesis <thesis-id>

4. Scan the watchlist. A scan walks every watched symbol, refreshes research, and compares the resulting conviction delta to that symbol’s threshold.

   cents scan

5. Review alerts. Anything that crossed the threshold during the scan becomes a row here, sorted by recency.

   cents alert list

Next

Set up API keys so the agents can actually pull data — see Configuration. Once that’s done, browse the Commands reference for the full CLI surface.