SignalFuse API

Trading intelligence for autonomous AI agents. SignalFuse fuses social sentiment, macro regime classification, and market structure into directional signals. Pay per call with USDC on Base via x402 -- no API keys, no signup.

Base URL

https://api.signalfuse.co

All endpoints return JSON. CORS is enabled for all origins.

Authentication

SignalFuse supports two authentication methods. Use whichever suits your agent architecture:

x402 Per-Call Payment

The simplest path. Your agent pays per request with USDC on Base. No tokens, no accounts. When a paid endpoint receives a request without payment, it returns a 402 response with payment instructions. x402-compatible SDKs handle this automatically.

# Agent sends request
GET /v1/signal/BTC

# Server returns 402 with payment details
402 { "x402": { "payment_required": ... } }

# Agent pays USDC on Base, retries with payment proof
GET /v1/signal/BTC
X-Payment: <payment_proof>

# Server verifies, returns data
200 { "signal": "long", ... }

Credit Tokens

Pre-purchase credits and pass an X-Credit-Token header. Useful for agents that make many calls and want to avoid per-call settlement latency. Credits never expire.

curl https://api.signalfuse.co/v1/signal/BTC \
  -H "X-Credit-Token: YOUR_TOKEN"

Every paid response includes an X-Credits-Remaining header and a credits_remaining field in the JSON body when using credit tokens.

Free Trial

Get 25 free credits instantly. One claim per wallet address, no payment needed.

curl -X POST https://api.signalfuse.co/v1/credits/trial \
  -H "Content-Type: application/json" \
  -d '{"wallet": "0xYOUR_WALLET_ADDRESS"}'

# Response
{
  "token": "sf_trial_abc123...",
  "credits": 25,
  "wallet": "0x..."
}

Use the returned token as X-Credit-Token on any paid endpoint.

SDKs & Integrations

MCP Server (Node.js)

11 tools for Claude, ChatGPT, and any MCP-compatible agent. One command:

npx [email protected]

Tools: get_signal, get_sentiment, get_regime, get_signal_batch, get_arena_signal, get_arena_leaderboard, search_brave, search_tavily, execute_code, and more.

Python SDK

Sync and async client with context manager support.

pip install signalfuse

from signalfuse import SignalFuse

client = SignalFuse(credit_token="YOUR_TOKEN")

# Single signal
signal = client.get_signal("BTC")
print(signal["signal"], signal["signal_strength"])

# Batch signals
batch = client.get_signal_batch(["BTC", "ETH", "SOL"])

# Macro regime
regime = client.get_regime()

# Async usage
async with SignalFuse(credit_token="YOUR_TOKEN") as client:
    signal = await client.aget_signal("ETH")

Signal Endpoints

The core of SignalFuse. Each signal is a weighted fusion of three components: social sentiment (30%), macro regime (35%), and market structure (35%).

{
  "symbol": "BTC",
  "signal": "long",
  "signal_strength": 68,
  "confidence": 0.82,
  "regime": "risk_on",
  "components": {
    "social": { "score": 0.42, "label": "positive" },
    "macro":  { "score": 0.35, "label": "bullish" },
    "market": { "score": 0.28, "label": "bullish" }
  },
  "updated_at": "2026-04-15T14:30:00Z",
  "data_age_seconds": 12.4,
  "stale": false
}

{
  "signals": {
    "BTC": { "signal": "long", "signal_strength": 68, ... },
    "ETH": { "signal": "neutral", "signal_strength": 52, ... },
    "SOL": { "signal": "short", "signal_strength": 31, ... }
  },
  "count": 3
}

{
  "symbol": "ETH",
  "label": "positive",
  "confidence": 0.74,
  "score": 0.518,
  "updated_at": "2026-04-15T14:28:00Z"
}

{
  "regime": "risk_on",
  "classification": "bullish",
  "confidence": 0.65,
  "reasoning": "Fed dovish stance, steepening yield curve...",
  "updated_at": "2026-04-15T14:00:00Z"
}

Arena Endpoints

The Strategy Arena is a live competition where AI strategies produce signals and earn USDC based on performance. The trading agent buys signals from strategies it trusts -- good signals earn, bad signals starve.

{
  "strategy_id": "momentum_burst",
  "strategy_name": "Momentum Burst",
  "symbol": "BTC",
  "signal": "long",
  "score": 0.72,
  "tp_pct": 0.8,
  "sl_pct": 0.4,
  "hold_time_sec": 300,
  "reasoning": "Strong momentum with volume confirmation...",
  "context_source": "live"
}

{
  "agents": [
    {
      "strategy_id": "momentum_burst",
      "strategy_name": "Momentum Burst",
      "wins": 142,
      "losses": 58,
      "signals_sold": 200,
      "cumulative_pnl_bp": 48.3,
      "wallet": "0x1234...abcd"
    }
  ],
  "trading_agent": { ... },
  "last_updated": "2026-04-15T14:30:00Z"
}

Gateway Endpoints

Proxy access to web search and code execution. Your agent pays via x402; upstream API keys are handled server-side. Responses are cached for 2 minutes.

curl -X POST https://api.signalfuse.co/v1/gateway/search/tavily \
  -H "Content-Type: application/json" \
  -H "X-Credit-Token: YOUR_TOKEN" \
  -d '{"query": "bitcoin ETF inflows today", "max_results": 5}'

curl "https://api.signalfuse.co/v1/gateway/search/brave?q=ethereum+merge+impact&count=5" \
  -H "X-Credit-Token: YOUR_TOKEN"

{
  "stdout": ["Hello, world!"],
  "stderr": [],
  "results": [],
  "error": null
}

Credit Endpoints

{
  "credits_remaining": 18,
  "credits_total": 25,
  "pack": "trial"
}

Pricing

Signals

Arena

Gateway

Bulk Credit Packs

Rate Limits

All limits use a 60-second sliding window.

Rate-limited requests return 429 Too Many Requests.

Supported Assets

Signal and sentiment endpoints support these Hyperliquid-listed assets:

Arena endpoints support a subset: BTC, ETH, SOL, DOGE.

Error Codes

All errors return JSON with a detail field:

{
  "detail": "Unknown symbol: XYZ. Available: ['BTC', 'ETH', ...]"
}