New Token Momentum Scanner — Early DEX Launches & Risk

Surface and rank just-created tokens across chains before they're on anyone's radar. The scanner pulls newly-created liquidity pools from multiple DEXes (keyless) and scores each on the observable early-momentum signals a person can't watch at scale — liquidity, volume acceleration, buy/sell pressure, unique-buyer growth, price action, turnover — alongside a risk/rug score (thin liquidity, FDV ≫ liquidity, sell pressure, too-new, few buyers). Read-only, keyless, agent-callable.

Signals, not predictions. This ranks observable on-chain/market activity. It is not a price forecast and not financial advice. Newly-created tokens are extremely high risk — most lose value or are scams/rugs. Use the risk flags, and never invest more than you can lose.

What it does

• Scans the newest pools on Ethereum, BSC, Base, Solana, Polygon, Arbitrum, Avalanche (keyless GeckoTerminal data).
• For each pool computes a transparent momentumScore (0–100) from: liquidity adequacy, volume acceleration (recent per-minute rate vs the hour's rate), buy pressure, unique buyers, price momentum, turnover.
• …and a riskScore (0–100) from: thin liquidity, FDV-to-liquidity ratio, sell pressure, too-few buyers, too-new-to-read.
• Returns the top candidates ranked by momentum, each with the raw numbers + human-readable reasons and risk flags + a GeckoTerminal link.

Field	Meaning
momentumScore / momentumLabel	0–100, quiet → building → active → heating → hot
riskScore / riskLabel	0–100, moderate → elevated → high → extreme
ageHours / stage	just-launched / early / establishing / maturing
liquidityUsd, fdvUsd, fdvToLiquidity	size & float signals
volumeH1Usd, volumeAcceleration	activity + whether it's accelerating
buysH1 / sellsH1 / buyPressure, uniqueBuyersH1	flow + organic interest
priceChangeH1Pct, turnover	price & volume-vs-liquidity
momentumReasons[] / riskFlags[]	why it ranked / what to watch

Why not just ask an AI agent?

An agent can look up one token. It can't continuously pull the newest pools across seven chains, compute volume acceleration and buy/sell pressure on multi-timeframe buckets, weigh FDV-against-liquidity rug signals, and rank hundreds of brand-new pools — deterministically, in seconds. That cross-source, at-scale scoring is the edge here; your judgment makes the call.

Input

Field	Type	Default	Description
networks	array	ethereum, bsc, base, solana	Chains to scan.
maxPagesPerNetwork	integer (1–5)	1	Pages of newest pools per chain (~20 each).
minLiquidityUsd	integer	3000	Ignore pools thinner than this.
maxAgeHours	integer	72	Only pools created within this window.
maxCandidates	integer	25	Top-ranked rows to return.
webhookUrl	string		POST the ranked scan here on completion.

Example input

{
  "networks": ["ethereum", "bsc", "base", "solana"],
  "minLiquidityUsd": 5000,
  "maxAgeHours": 48,
  "maxCandidates": 25
}

Output

One row per candidate, ranked by momentum. Example (real run, trimmed):

{
  "network": "bsc",
  "dex": "pancakeswap_v2",
  "name": "交易猫 / WBNB",
  "ageHours": 0.08,
  "stage": "just-launched",
  "liquidityUsd": 39974.8,
  "fdvUsd": 153217.62,
  "fdvToLiquidity": 3.8,
  "volumeH1Usd": 23913.51,
  "volumeAcceleration": 3.14,
  "buysH1": 81, "sellsH1": 65, "buyPressure": 0.555,
  "uniqueBuyersH1": 58,
  "priceChangeH1Pct": 140.35,
  "momentumScore": 72, "momentumLabel": "hot",
  "riskScore": 12, "riskLabel": "moderate",
  "momentumReasons": ["volume accelerating (recent rate 3.14x the hourly rate)", "58 unique buyers in the last hour", "price +140% (1h)"],
  "riskFlags": ["too new to read (<15 min)"],
  "geckoTerminalUrl": "https://www.geckoterminal.com/bsc/pools/0x95a8..."
}

Use cases

• Early-mover discovery — find pools gaining real, accelerating activity in their first hours.
• Rug avoidance — the risk score + flags surface thin-liquidity / thin-float / sell-heavy launches to avoid.
• A standing feed — schedule + webhook for a continuous "new & heating" stream to Slack/Discord.
• AI trading/research agents — a ranked, scored candidate list to feed into the agent's own analysis.

Alerts / scheduling

Set webhookUrl and schedule the actor (e.g. every 15–30 min) — each run POSTs the freshly-ranked candidates to Make/Zapier/your endpoint, so you see new heating launches as they appear.

For AI agents / MCP

https://mcp.apify.com/?tools=actors,docs,centorprat/new-token-momentum-scanner

Loads as centorprat--new-token-momentum-scanner. Claude Code:

claude mcp add apify -- npx -y @apify/actors-mcp-server --actors centorprat/new-token-momentum-scanner

FAQ

Does it predict which token will pump? No. It ranks current observable signals (liquidity, accelerating volume, buy pressure, buyers, price). It does not and cannot predict future price — treat it as a discovery + due-diligence aid, not advice.

Which chains? Ethereum, BSC, Base, Solana, Polygon, Arbitrum, Avalanche.

Does it need an API key? No — keyless public DEX market data.

How does it handle scams/rugs? It can't check contract ownership or liquidity locks, but it surfaces the observable risk signals (thin liquidity, FDV ≫ liquidity, sell pressure, too-new, almost no buyers) as riskFlags so you can avoid the obvious traps. Always verify the contract yourself.

What is "volume acceleration"? The recent per-minute volume rate (last 5 min) divided by the hour's per-minute rate — above 1 means activity is speeding up.

How fast / how much? A couple of seconds per scan. Pay-per-event: a base per scan + a small premium per candidate returned.

Data & attribution

Public DEX market data via the keyless GeckoTerminal API. No keys, no personal data. All outputs are computed signals from public data, not financial advice.

Related actors

• Bitcoin Wallet & Transaction Tracer
• Ethereum Wallet & Token Tracer
• ERC-20 Token Intelligence & Holder Concentration
• Crypto Address Risk & AML Screener