DeFi Safety Oracle

Focused honeypot check: a live buy-then-sell simulation against real DEX liquidity (Aerodrome, Uniswap V2/V3/V4). Tells an agent whether the token can actually be SOLD, the round-trip loss percent, and whether GoPlus or honeypot.is flag it too. Catches sell-blocking scam tokens, extreme sell taxes and no-liquidity traps before a buy is placed.

Token contract risk check for any EVM token, run by us directly on chain: rug-pull red flags such as unverified source, code that can be swapped later (upgradeable contract) and who controls that, ownership not renounced, and dangerous powers in the bytecode (mint, pause, blacklist, fee-setting). Returns the raw on-chain evidence, not just a score, so an agent can justify its decision.

Cheapest identity taster ($0.001): feed it a 0x address or @handle, get back the input kind, a trust/credibility verdict, the top 3 red-flag signals, a one-line summary and an upgrade URL to the full /v1/identity/token or /v1/identity/x endpoint. Uses cached full results when present, else the cheapest reads only. Billable only when it answered.

Who is behind a token: our safety verdict (grade, honeypot), DexScreener market (liquidity, volume, age, pair), declared X + website with contract/X linking checks, the creator wallet (Clanker admin or on-chain deployer) profiled as /identity/wallet, stable red-flag signals (impersonation, wash trading, dead token, serial rugger, rug ring) and a credibility verdict. Not charged when nothing could be read.

Is this X (Twitter) account real? Age, followers, following, post count, verification, website and contract addresses in the bio, with red-flag signals: brand-new account, bought-looking following, follow-spam, deleted/suspended. For vetting token projects, KOLs and support impostors. Returns signals plus a trust verdict.

Who is behind a wallet: age, activity, who first funded it, its ENS/Basename and the X account linked on it, every token it deployed and how many we proved were honeypots or scams (from 130k graded Base tokens), rug-ring funders and address-poisoning attempts. Returns stable signal ids plus a trust verdict. Not charged when the wallet has no history.

Crowd Check instant verdict for trading bots ($0.01): ok, caution or avoid for a token, with risk score, top reasons and key flags (liquidity pullable, serial launcher, v4 hook, bundled share). Tokens our hunt already checked answer instantly; others return 202 with no charge until ready. Base, Ethereum, Robinhood Chain, Solana. ?chain=&token=

Paper test ($0.05): what would have happened if you had followed a rule on every new token we found? Pick a preset (avoid_farms, locked_pools_only, clean_checks, small_pools_only, farms_only, everything) or set your own filters, a horizon of 1, 6 or 24 hours and a bankroll. Returns trades, win rate with interval, average and median return, drawdown, crash rate, and the same numbers for ALL tokens as a baseline, with fees and price impact charged.

Crowd Check creator check ($0.01): who is behind a token - its creator wallet, the funding chain behind it (up to 8 hops), the track record of earlier tokens, wallet age and throwaway-wallet signals, and any registered name and X account linked to the creator (claimed or proven both ways). Serial launchers flagged. Base, Ethereum, Robinhood Chain. Free quote first at /v1/holders/quote, then ?quote=<id>.

Crowd Check pool safety ($0.01): can the token's liquidity be pulled - who holds the pool share (burned, wallet, or a lock service with the real locked share and unlock date) and, for Uniswap v4, what the pool's hook may do on every trade and whether it is a launch platform's shared hook. Free quote first at /v1/holders/quote, then ?quote=<id>.

How old is this account? Pass ONE of x=@handle, wallet=0x... (chain=base|ethereum) or tg=<numeric Telegram user id>. Returns created_at, age_days and the method: exact for X (profile join date) and wallets (first on-chain activity), an ESTIMATE with an earliest/latest range for Telegram. For fresh-account and sybil filters. Not charged when nothing is found.

Crowd Check quick check ($0.02): instant risk score 0-100 with reasons for a token on Base, Ethereum, Robinhood Chain or Solana - snipers, launch bundles (wallets fed in one launch tx), dev holding and whether the dev sold, dev-fed insiders, padded holder count, top-10 share, whether the creator can pull the pool (liquidity lock), risky trading hooks, creator rug record. Get a free quote at /v1/holders/quote, then ?quote=<id>.

One-call launch screening for launchpads and trading apps: pass a token, get a GREEN / CAUTION / BLOCK decision (action: allow / review / reject) with plain-English reasons. Combines a live honeypot/tax simulation and owner-power analysis with the checks no other scanner has — whether the DEPLOYER or the exact CODE has rugged before (from a 35k-verdict ledger). Drop it into your listing flow to auto-reject rugs before they ever go live. Cache-first, so a repeat token is near-instant and free.

Crowd Check full scan ($1): is the crowd real? Profiles the top 300 real holders of a token on Base, Ethereum, Robinhood Chain or Solana - wallet age, real funder traced through routers, funder clusters, fresh wallets, snipers, launch bundles, dev and insider holdings, risk score 0-100, verdict organic/mixed/farmed. Free quote first at /v1/holders/quote, then ?quote=<id>; returns a job, call again with the same quote for the result.

Watch a token for 24 hours ($0.10). Returns a ticket. Poll it later for alerts: liquidity down 30 percent, price down 50 percent, top holder up 10 points, creator balance down 50 percent. Plain language, one sentence each. Pay once, then poll as often as you like at a few cents a poll.

Deployer (creator-wallet) reputation from our own 35k-verdict ledger. Pass a token and we resolve its deployer, or pass a wallet directly. Returns how many of that wallet's prior tokens we proved were honeypots or scams (realised harm only; tokens that merely died with no liquidity are reported as context, never counted as rugs), the malice rate, a rug-risk penalty and reason codes. A prior to look harder or price wider, never proof this token is a honeypot.

The plain-language card for a token ($0.10): who holds the supply, who can pull the pool, who holds powers over the token, and the largest sensible position, in under 900 characters, as text plus a verdict (go, caution, no_go). Free quote first at /v1/holders/quote, then ?quote=<id>. Not financial advice.

Ask a watch what changed ($0.01). Pass the ticket from /v1/watch/start. Compares the token now with the last poll and returns plain-language alerts, if any. Not charged if the market data is unavailable or the watch has expired.

Full safety report for an ERC-20 token. Runs a live buy-then-sell honeypot and tax simulation against real DEX liquidity (Aerodrome, Uniswap V2/V3/V4), scans contract powers (upgradeable proxy, mint, pause, blacklist, fee-setting), resolves ownership and effective controllers, and cross-checks GoPlus and honeypot.is. Returns an A-F grade, tradeable flag, 0-100 risk score and evidence-carrying flags an agent can cite.

Full multi-dimensional token risk decomposition for agent decisioning: separate 0-100 scores for honeypot, rug pull, admin abuse and liquidity risk, plus confidence, honeypot and rug probabilities, buy/sell tradeability with sell tax in basis points, contract privileges (mint, pause, blacklist, upgrade, tax change), effective controllers, dynamic-tax caps, and a buy/sell simulation matrix that detects conditional, size and wallet traps.

Batch token safety screen: submit up to 25 tokens in one call and get each one's A-F grade, tradeable flag, honeypot verdict and risk flags, plus a summary count of tradeable tokens, honeypots and errors. Runs the same engine as the single-token report, in parallel, with a per-token timeout so one bad token cannot block the batch. Use it to filter a watchlist or a new-pair feed.