Stop duct-taping five data subscriptions together. CoinGlass + Santiment + DEX data + an LLM + liquidation ML costs $600+/month and two weekends of integration.
SignalHub fuses all of it into a single 0–100 score per pair — and exposes every component so you can see inside the number.
$ curl https://signalhub.io/api/v1/signals/latest/BTCUSDC \ -H "Authorization: Bearer sh_..." { "pair": "BTCUSDC", "score": 72, "signal": "BULLISH", "components": [ { "name": "openmarket_regime", "raw": 80, "weight": 0.30 }, { "name": "openmarket_pair", "raw": 65, "weight": 0.20 }, { "name": "breakout_signals", "raw": 70, "weight": 0.15 }, { "name": "liquidation_risk", "raw": 65, "weight": 0.15 }, { "name": "dex_arb_spread", "raw": 60, "weight": 0.10 }, { "name": "sentiment", "raw": 55, "weight": 0.10 } ] }
Public endpoint returns score + signal · Free account unlocks all 6 component scores
Six live data sources, each with a defined weight and an inspectable raw score. You can see exactly why BTC scores 72 right now.
An LLM analyzes live market microstructure every 5 minutes, classifying the regime: trending_up, ranging, trending_down, volatile, or halt. Output maps to a 0–100 score.
Per-pair buy_scale from a grid bot with real capital on Binance. When the bot actively scales into a position, the score reflects it. Skin in the game as a live signal.
Rolling win-rate from a live breakout strategy plus active entry signals per pair. Win-rate above 60% lifts the score; below 40% suppresses it. Updates with each new trade close.
ML model predicting cascading liquidation probability from on-chain sequences. High liquidation risk suppresses the score; low risk is a bullish indicator. Inverted intentionally.
Average arb spread in basis points across DEX/CEX venues. Tight spreads (liquid, healthy market) score high. Wide spreads (dislocated, volatile) score low.
Market sentiment from LLM-extracted Fear & Greed reasoning, blended with altcoin social signals. Provides the macro emotional context for each 5-minute cycle.
The buy_scale input is unique. It comes from a grid bot with real capital deployed on Binance — not a backtest, not a simulated portfolio. When the bot actively scales into positions, the score rises. No other data vendor has access to live bot operational telemetry as a signal source.
Building the same signal stack yourself costs $600+/month — before engineering time.
| Product | Composite score | Inspectable | 5-min updates | Bot telemetry | Dev entry price |
|---|---|---|---|---|---|
| SignalHub | Free–$49/mo | ||||
| Token Metrics | $99/mo | ||||
| altFINS | $299/mo | ||||
| CoinGlass | $29/mo | ||||
| CryptoQuant | $39/mo | ||||
| Santiment | $49/mo | ||||
| Glassnode | $800+/mo |
Gate entries on score threshold. Poll the endpoint in your strategy loop. If score > 70 and regime is trending_up, scale in. Add webhooks for push notifications instead of polling.
score >= 70 && regime === "trending_up" → enterRebalance crypto exposure based on aggregate score across all pairs. When both BTC and ETH drop below 35, reduce risk. React to regime transitions without watching charts.
avg_score < 35 → reduce_crypto_exposure()Pull historical scores via /signals/history and correlate against 24h forward returns. Validate that score > 75 has positive expectancy before deploying capital.
GET /signals/history?pair=BTCUSDC&from=2026-01-01No rate-limiting surprises. No enterprise gating. Real prices, published.
Validate before you commit
For active bot developers
For production systems
All plans include rate-limit headers (X-RateLimit-*), CORS, and standard JSON. No hidden fees.
Free API key, no credit card. One request to validate it works before you commit to anything.