name: pumpfun-sniper description: | Score any pump.fun token CA for snipe safety (0-100) before buying. Analyzes dev wallet history, social links, liquidity, and holder concentration. Returns SNIPE / CAUTION / AVOID verdict with detailed signal breakdown. Use when the user wants to snipe a new token, check if a pump.fun token is safe, score a new launch, detect rug potential, or analyze a fresh CA before buying. Keywords: pump.fun snipe, rug check, new token score, CA analysis, safe to buy. metadata: openclaw: requires: bins: ["python3"] env: ["HELIUS_API_KEY"]

pump.fun Sniper Scorer

Score any Solana token CA for snipe safety before you buy. Returns a 0–100 score with verdict: SNIPE / CAUTION / AVOID.

Paid API (Recommended)

Hosted endpoint — $0.05 USDC per request via x402 on Base chain:

# Check payment requirements
npx awal@latest x402 details https://pumpfun-sniper-production.up.railway.app/score?ca=TOKEN_CA

# Score a token (auto-pays)
npx awal@latest x402 pay "https://pumpfun-sniper-production.up.railway.app/score?ca=TOKEN_CA"

What Gets Scored

Verdict Guide

API Response

{
  "ca": "TokenCA...",
  "score": 74,
  "verdict": "SNIPE",
  "token": {
    "symbol": "PEPE",
    "name": "Pepe Coin",
    "price_usd": "0.0000123",
    "liquidity_usd": 45000,
    "market_cap": 180000,
    "twitter": "https://twitter.com/pepecoin",
    "telegram": "https://t.me/pepecoin"
  },
  "breakdown": {
    "dev_wallet":    {"score": 22, "max": 30},
    "social":        {"score": 18, "max": 20},
    "liquidity":     {"score": 20, "max": 25},
    "concentration": {"score": 14, "max": 25}
  },
  "signals": [
    "✅ Dev has significant SOL (12.5 SOL)",
    "✅ Token: $PEPE — Pepe Coin",
    "✅ Has description",
    "✅ Twitter: https://twitter.com/pepecoin",
    "✅ Strong liquidity: $45,000",
    "✅ Strong buy pressure: 45B / 12S (5m)",
    "⚠️ Top 5 hold 38.2%"
  ]
}

Agent Usage

# Self-hosted (requires HELIUS_API_KEY for best results)
python3 {baseDir}/scripts/scorer.py <TOKEN_CA>

Always use --json / the API endpoint for agent use. Scoring takes 10–20 seconds due to multiple on-chain lookups.