A hands-on track for building with the TradeWave Data API and MCP server. Every lesson uses runnable code and the live v1 endpoints. New here? Start with lesson 1, or jump straight to your first API call.
Following the REST API lessons? Create a developer key in the
dashboard and send it as
Authorization: Bearer tw_live_.... The MCP setup lesson uses the
TradeWave account authorization shown in that guide.
What is a seasonal edge, and how to read a TradeWave Pattern Card
Learn what a seasonal edge is and how to read every field of a TradeWave Pattern Card, from edge_score to receipts and the broker-agnostic order ticket.
Your first TradeWave API call in 5 minutes
Make your first TradeWave call with the public demo token, then swap in your own key and read GET /daily-pick.
Build a cross-market seasonal screener with /scan
Use the flagship GET /scan endpoint to rank seasonal opportunities across every market in your scope, with honest neutral-bias filtering.
Using the ML win-probability model
Read the ml block on a TradeWave Pattern Card, tell ml_win_prob apart from historical_win_rate, and handle metering and the 90-day horizon gracefully.
Connect TradeWave to ChatGPT or Claude
Add the TradeWave server, sign in to your TradeWave account, and start asking market-research questions.
The verifiable track record
Why a time-stamped, forward-tested daily-pick ledger is TradeWave's core trust asset, and how to read it with GET /daily-pick.
Rate limits, quotas, and tiers in practice
Read the X-RateLimit headers, back off on 429 and 5xx, handle the ML upgrade nudge, and manage keys in the TradeWave console.
Recipe: TradeWave + your broker (works with any execution app)
A cookbook flow: ask TradeWave for the best seasonal setup, then place the broker-agnostic order ticket at any broker. We never take your trades.
Build, do not rebuild: why your coding agent should consume TradeWave, not recompute it
An AI agent can write a seasonality backtest in minutes. Here are the four reasons that backtest will not be tradeable, and what to consume instead.
What you will be able to do
- Read a Pattern Card and trust it (receipts, per-year results,
neutralhonesty). - Scan every market you can trade and rank the best seasonal setups with one call.
- Use the ML win-probability model, and handle its coverage gracefully.
- Connect Claude, Cursor, or ChatGPT to TradeWave over MCP.
- Build robust clients (rate limits, quotas, retries) and place trades at any broker.