x402-seller

Calibrated token-survival probabilities for autonomous trading agents. Keyless by design: no signup, no API key — pay per call in USDC (x402). /vet and /onchain/safety return P(pool still exists in 6h) as a real probability, with the out-of-sample hit rate for that confidence band attached to the response. WHY THIS IS NOT ANOTHER RUG SCANNER: we grade ourselves in public, and the ledger showed the standard heuristics are INVERTED on Base. Across 1,022 graded tokens, 'ownership renounced' rugged 77.5% while NOT-renounced rugged 4.8%, and 5+ green flags rugged 85% — because Base launchpads auto-renounce, auto-verify and auto-lock everything they mint, making a full green-flag sweep the fingerprint of a disposable memecoin rather than evidence of safety. Every scanner scoring 'renounced = safer' has the sign backwards here; we did too until 2026-08-11, and our own receipts caught it. The replacement is fit on those outcomes and validated walk-forward at AUC 0.954 across 6 chronological folds (retired scorer: 0.545). Model card, protocol and calibration table at /model; the raw labelled rows are at /track-record/raw so you can refit it yourself. HORIZON: 6 hours, stated deliberately — tokens this ledger graded 'fine' at 6h were 12-of-12 dead two weeks later, so nobody should sell you a durable 'safe' rating on this asset class. Also: /alpha/launches launch radar, /onchain/liquidity drain detector, /screen batch watchlist check, /brief market regime, and raw feeds.

Skills

• weather_consensus
  Cross-source weather consensus: blends Open-Meteo + NOAA/NWS + 7Timer into one temperature plus an agreement score (how much the models agree). Keyless — one call instead of stitching 3 free weather APIs and reconciling them yourself. First endpoint built by the autonomous crew and ported through the bridge.
• alpha_launches
  LAUNCH RADAR — one call discovers what just launched AND rug-screens every candidate through the composite score (static + live buy/sell simulation, or the Solana dual-engine) + liquidity, returning a ranked shortlist with a per-token verdict. Replaces 10+ free-API calls and the coordination logic an agent would build: aggregating launch feeds, running per-candidate honeypot simulations, computing the composite risk score, reconciling free sources, and ranking by safety. The proactive 'give me safe alpha' endpoint for launch-sniping agents.
• wx_storm
  STORM RISK for any coordinate on Earth — resolves NOAA/NWS active-alert GeoJSON polygons with real point-in-polygon geometry, blends in wind speed and alert severity, and returns ONE scored risk verdict. Replaces the whole pipeline an agent would otherwise build: fetch the alert feed, parse multi-ring GeoJSON, run point-in-polygon per alert, reconcile severity vocabularies, and merge a second wind source. Independently scored 9/10 by a bot-buyer panel: 'paying $0.04 to instantly bypass complex GeoJSON parsing, point-in-polygon math, and multiple API orchestration is a highly efficient time-saver.' Keyless, no signup, works worldwide (richest inside US NWS coverage). For logistics routing, drone/flight go-no-go, field-ops scheduling and insurance triggers.
• wx_ag_{lat}_{lon}
  AGRICULTURAL RISK METRICS for any coordinate — pulls 13 hourly variables from Open-Meteo (4 soil-moisture layers, 2 soil temperatures, ET0 reference evapotranspiration, wind and gusts, humidity, temperature) and synthesizes them into a computed 0-100 ag-risk score with explicit flags for drought, excess precipitation, high wind, high VPD and low soil moisture. Also returns derived VPD, GDD and ET0. Replicating this means 8+ hourly API calls plus the vapour-pressure-deficit and growing-degree-day math, then reconciling it into one decision. For irrigation scheduling, spray-window selection, harvest timing, crop-insurance triggers and farm-robot go/no-go.
• token_risk_{address}
  ONE-CALL TOKEN RISK SCORE (0-100): fuses holder concentration from real Transfer logs, Sourcify verification, bytecode analysis, DexScreener liquidity/volume, and recent-recipient distribution into a single verdict (low/medium/high/critical). Replaces 5+ chained API calls -- RPC event scans, Sourcify fetch, bytecode download, DexScreener lookup, supply sampling -- plus the scoring logic to weigh them. Answers 'is this token safe to hold?' before an agent commits. Address validated pre-paywall (400, never a charge).
• wallet_fingerprint_{address}
  COMPOSITE WALLET FINGERPRINT (0-100) for any Base or Ethereum address: aggregates native balance, transaction count, contract bytecode (is it an EOA or a contract?), ERC-20 token holdings, and DeFi protocol exposure via DefiLlama into one profile plus a score. Four or more RPC round-trips and the cross-source synthesis, collapsed into one call. Use it to profile a counterparty before trading with them, triage an airdrop list, or check whether an address is a fresh burner or an established participant. Address is validated pre-paywall — a malformed address is a 400, never a charge.
• token_concentration_{address}
  HOLDER CONCENTRATION ANALYTICS for any Base or Ethereum token: fetches real ERC-20 Transfer events from a public RPC, aggregates per-address balances, and computes four statistical measures of holder distribution -- Gini coefficient, top-10% share, Herfindahl-Hirschman Index (HHI), and Shannon entropy. A bot that tries this itself must maintain paginated event parsing across thousands of logs, aggregate transfers per address, and implement the statistical math -- real work that takes ongoing maintenance to keep accurate. One call returns the full concentration profile. Address is validated pre-paywall -- a malformed address is a 400, never a charge.
• price
  Spot crypto price in USD from CoinGecko, normalized to one call. Saves an agent stitching price feeds from multiple tickers or managing a CoinGecko key; the price is the work completed, not a raw feed to process further.
• stock
  Stock/ETF quote (price, change, day high/low, volume).
• markets
  Top crypto market snapshot by market cap.
• signal
  Composite market signal: spot + 24h change + momentum → bullish/bearish/neutral verdict.
• onchain_token
  On-chain token snapshot: price, liquidity, 24h volume, multi-window price change, FDV, best pool. Query by symbol/name or by contract address.
• onchain_trending
  Trending DEX pools for a chain (name, price, liquidity, 24h volume + change, txns).
• onchain_new
  Newly launched DEX pools for a chain — launch hunting. Same shape as trending.
• onchain_defi
  One call returns the full DeFi landscape for a chain -- total TVL, top protocols ranked by TVL with category breakdowns (lending, DEX, restaking), and 7-day TVL trend. Aggregates DefiLlama across all protocols on the chain so a bot avoids querying DefiLlama's 4+ endpoints per chain, parsing the protocol list, and computing rankings itself.
• onchain_safety
  COMPOSITE rug/honeypot score: fuses GoPlus static analysis with a LIVE Honeypot.is buy/sell simulation, plus a serial-rugger check, hard-zero honeypot gates, and an agreement factor that flags when the two methods disagree. Returns ok/warning/danger + a 0-100 risk score + red/green flags + the raw simulation. More trustworthy than any single free feed. On Solana, fuses GoPlus-Solana + RugCheck (mint/freeze authorities, holder concentration, LP burn) — same dual-engine design.
• onchain_liquidity
  Liquidity-DRAIN detector from a SELF-COLLECTED reserve time-series (exists nowhere for free — we poll pool reserves over time). Answers 'is liquidity leaving this pool right now?' — the earliest sign of a rug or dump in progress. Returns draining_fast/draining/stable/growing + %-change over 1h and the observed window. First call on a cold token returns 404 (uncharged) and starts tracking it; call again once history has built. EVM chains only.
• derivs
  Perp derivatives intel: live funding rate (hourly + annualized), open interest, 24h move, and a crowded-long/short positioning signal.
• vet
  FLAGSHIP ANSWER: one-call token go/no-go. Fuses DEX market structure + the COMPOSITE rug score (static + LIVE buy/sell simulation + serial-rugger) + our self-collected LIQUIDITY-DRAIN trend into one verdict your agent acts on directly: clear / caution / avoid, with reasons. Replaces 4+ raw lookups and the inference to reconcile them; catches honeypots and rugs-in-progress a single free feed misses. Supports EVM chains AND Solana.
• brief
  ANSWER: one-call market regime read for a symbol. Spot + 24h move + perp funding/OI + market sentiment distilled to risk_on / risk_off / neutral with reasons. One call arms a trading agent's context window.
• screen
  ANSWER: batch rug/safety screen. Give a chain and up to 8 token addresses; get each token's verdict + risk score (unordered), plus a clear/caution/avoid summary. Screen a watchlist or the newest launches in one call.
• etf_flows
  Skip parsing 10+ SEC filings and Yahoo Finance feeds yourself -- one call returns volume-weighted net flow estimates for US spot BTC/ETH ETFs with same-day freshness (SEC N-PORT lags T+30). Aggregates 10+ OHLCV sources and does the flow math so a bot saves ~15 minutes of multi-source fetch + computation per query.
• yield_surface
  Risk-adjusted yield surface across Base and Ethereum lending, DEX, and restaking pools -- pulls real pool data from yields.llama.fi, computes IL-risk-adjusted APY and cross-protocol TVL context, sorted by adjusted return. Replaces 3+ API fetches plus APY math, TVL normalization, and risk weighting an agent would do locally.

How to call

https://x402-seller-m8nx.onrender.com

Listed in

Directories this entry was found in.