Three workflows worth copying.

All three reuse one paid client. Set it up once:

import os
from x402.clients.requests import x402_requests
from eth_account import Account

apex = x402_requests(Account.from_key(os.environ["EVM_PRIVATE_KEY"]))

def sig(name: str) -> dict:
    return apex.get(f"https://apexrunner.ai/signals/{name}").json()

Workflow 01 — DCA timing agent

Use case: decide whether to add to a spot position, only when several independent conditions line up — so you accumulate into weakness instead of chasing.

Signals: dca-reentry-gate · fear-greed · regime-confluence

Logic: add only when the re-entry gate reads OPEN, sentiment is fearful (a contrarian entry condition), and the regime is ranging or neutral rather than trending or in crisis. Any one failing is a pass.

def should_dca() -> bool:
    gate   = sig("dca-reentry-gate")
    fg     = sig("fear-greed")
    regime = sig("regime-confluence")

    return (
        gate["status"] == "OPEN"
        and fg["value"] <= 25
        and regime["regime"] in ("ranging", "neutral")
    )

if should_dca():
    place_dca_order()   # your own execution layer

Workflow 02 — Risk management agent

Use case: decide whether to trim or hedge an open position before the market does it for you, by reading crowding and stress rather than price alone.

Signals: crowded-trade-detector · liquidation-pressure · portfolio-heat

Logic: raise a reduce-risk flag when a position is crowded and liquidation pressure is building and your own book is running hot. Crowding alone is noise; the combination is a setup for a cascade.

def risk_action(symbol: str) -> str:
    crowd = sig("crowded-trade-detector")
    liq   = sig("liquidation-pressure")
    heat  = sig("portfolio-heat")

    crowded   = crowd["crowding_score"] >= 70        # 0-100, higher = more crowded
    squeezing = liq["pressure"] == "elevated"
    hot       = heat["heat_pct"] >= 8               # your book's risk utilisation

    if crowded and squeezing and hot:
        return "REDUCE"      # trim or hedge
    if crowded and squeezing:
        return "WATCH"       # tighten stops, no new size
    return "HOLD"

Workflow 03 — Execution optimizer

Use case: you've decided to trade; now place it well — choose the venue with the best expected fill and wait for a low-slippage window instead of crossing the spread on impulse.

Signals: optimal-order-routing · slippage-forecast · execution-window-optimizer

Logic: route to the venue the router favours, but only submit when the forecast slippage is acceptable and the window optimizer says conditions are favourable; otherwise hold and re-check.

def plan_execution(symbol: str, notional_usd: float) -> dict:
    route  = sig("optimal-order-routing")
    slip   = sig("slippage-forecast")
    window = sig("execution-window-optimizer")

    ok_to_send = (
        slip["expected_bps"] <= 15            # cap acceptable slippage
        and window["recommendation"] == "execute"
    )

    return {
        "venue": route["best_venue"],
        "send_now": ok_to_send,
        "expected_slippage_bps": slip["expected_bps"],
    }

Cost

A workflow costs the sum of its signals' list prices minus your wallet's discount tier (30% during the early-adopter window). Prices and discounts change, so rather than budgeting against a fixed total, call my-pricing for the exact amount a wallet will be charged:

curl https://apexrunner.ai/signals/my-pricing