Build on the signal panel

Authentication

Create a key at /account/api-keys and send it as a bearer token. The signal endpoint accepts any key (non-Quant keys get a 90-day-delayed view); the backtest endpoint requires a Quant key.

curl https://www.tradingagentapp.com/api/v1/signals?market=US&horizon=7d&limit=5 \
  -H "Authorization: Bearer ta_live_xxxxxxxx"

No account yet? Use the public demo key — literally the word demo. It works on every endpoint at free-tier limits (fundamentals: 25 flagship tickers at full fidelity; signals: 90-day-delayed view). Zero signup, zero card.

GET /api/v1/fundamentals

Point-in-time company fundamentals from official regulator filings — US SEC EDGAR (annual + quarterly), Taiwan TWSE (all listed companies), Korea OpenDART. Every figure is as first reported, keyed by its filing date: query with as_of and you see only what was public that day. Pricing and licences: /fundamentals.

# works right now — public demo key, no signup
curl "https://www.tradingagentapp.com/api/v1/fundamentals?ticker=2330.TW" \
  -H "Authorization: Bearer demo"

# the as-of snapshot a backtest at 2022-01-01 was allowed to see
curl "https://www.tradingagentapp.com/api/v1/fundamentals?ticker=AAPL&as_of=2022-01-01&view=latest" \
  -H "Authorization: Bearer demo"

Row fields: m metric · fy/fp fiscal year/period (FY, Q1–Q4) · end period end · filed filing date (the PIT key) · v/u value/unit. Metrics span raw statement lines (revenue, net_income, assets, equity, eps, …) and derived ratios (net_margin, roe, roa, debt_to_equity, …). Notes: Taiwan income-statement figures are cumulative year-to-date (Taiwan convention — difference adjacent quarters for single-quarter flows); every response embeds its statutory source attribution.

GET /api/v1/signals

The resolved signal panel — one row per name per date, with the model's forecast and the realised forward return. Resolved/ historical only (a factual track record, safe in every jurisdiction); never forward recommendations.

{
  "object": "signal_list",
  "tier": "studio",
  "full_access": true,
  "delayed_days": 0,
  "count": 5000,
  "schema": { "predicted_pct": "forward-return forecast (fraction)", "...": "..." },
  "data": [
    { "date": "2025-05-20", "ticker": "AAPL", "market": "US", "horizon": "7d",
      "predicted_pct": 0.0257, "actual_pct": 0.0154, "status": "won", "source": "live" }
  ]
}

POST /api/v1/backtest

Run a cross-sectional strategy and get a costed result: equity curve, Sharpe/IR, turnover, walk-forward out-of-sample folds — or a parameter sweep with the overfit-corrected Deflated Sharpe. Backtest our archive (source) or your rows (panel). Quant key required.

curl -X POST https://www.tradingagentapp.com/api/v1/backtest \
  -H "Authorization: Bearer ta_live_xxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "config": { "direction": "long_short", "quantile": 0.1, "weighting": "equal", "holdDays": 21 },
    "source": { "market": "TW", "horizon": "7d" },
    "sweep": false
  }'

{
  "object": "backtest_result",
  "n_rows": 8421,
  "metrics": {
    "cagr": 0.123, "sharpe": 1.42, "lo_sharpe": 1.31, "information_ratio": 0.88,
    "max_drawdown": 0.19, "avg_turnover": 0.34, "cost_drag_annual": 0.021,
    "oos_sharpe_mean": 0.97
  },
  "equity_curve": [1, 1.01, 0.99, ...],
  "folds": [ { "fold": 0, "sharpe": 1.1, "cagr": 0.14, "...": "..." } ]
}

GET /api/v1/fundamentals

Point-in-time company fundamentals from official regulator filings — US live (SEC EDGAR), more markets added as their official source allows (Korea via FSS OpenDART). As-first-reported — every row carries its official filed date, and queries are filtered to filed ≤ as_of, so a backtest at date T sees only what was public at T (no restatement look-ahead). Annual raw concepts + derived ratios (net_margin, roe, roa, debt_to_equity, asset_turnover…). Omit ticker to list available markets + tickers.

curl "https://www.tradingagentapp.com/api/v1/fundamentals?ticker=AAPL&metric=net_margin,roe&as_of=2022-01-01" \
  -H "Authorization: Bearer ta_live_xxxxxxxx"
# → only 10-Ks filed on/before 2022-01-01 (point-in-time)

Python — pip install pit-fundamentals

Zero-dependency package (pandas optional) with the demo key built in:

pip install pit-fundamentals

import pit_fundamentals as pf
rows = pf.fundamentals("2330.TW")                 # works instantly — demo key default
snap = pf.latest("AAPL", as_of="2022-01-01")      # the PIT snapshot
panel = pf.signals(market="US", horizon="7d")     # scored history (delayed on free)
# paid key: export PIT_FUNDAMENTALS_KEY=ta_live_…  (code unchanged)

The original single-file SDK also remains: Download tradingagent.py.

from tradingagent import TradingAgent
ta = TradingAgent()  # reads TRADING_AGENT_API_KEY

sig = ta.signals(market="TW", horizon="7d", limit=5000)

bt = ta.backtest(
    config={"direction": "long_short", "quantile": 0.1, "holdDays": 7},
    source={"market": "TW", "horizon": "7d"},
)
m = bt["metrics"]
print(f"Sharpe {m['sharpe']:.2f}  CAGR {m['cagr']:.1%}  maxDD {m['max_drawdown']:.1%}")

# Overfit-corrected sweep
sw = ta.backtest(config={"holdDays": 7}, source={"market": "US", "horizon": "7d"}, sweep=True)
print(sw["best"]["label"], sw["deflated_sharpe"]["dsr"], sw["deflated_sharpe"]["survives"])

Notebooks — runnable on the demo key

Three worked examples, each runnable end-to-end with zero signup (open in Jupyter, VS Code, or Colab):

• The reporting-lag trap — why joining fundamentals on period end (not filing date) silently inflates backtests — measure the lag yourself
• Taiwan fundamentals in 5 minutes — TSMC point-in-time, plus the cumulative-YTD convention every TW quant must know
• Fundamentals meet a scored signal panel — join PIT factors to realised forward returns with zero leakage by construction