Seven analyzers.
One honest score.
Every token that lands in your panel has been evaluated across 17+ risk vectors, grouped into seven analyzer modules. No secret sauce — just deterministic rules you can inspect at /methodology.
Can the deployer still screw you?
Mint function, freeze authority, blacklist capability, proxy upgrade path, ownership status, transaction tax. Seven different ways a smart contract can turn against its holders — all checked the moment the pool opens.
-
has_mint_functionSupply can be inflated-20
-
has_freeze_functionTransfers can be frozen-15
-
ownership_not_renouncedDeployer still in control-15
-
is_proxyLogic can be swapped-10
-
high_taxHoneypot risk-5 to -25
How deep, how locked, how real?
Pool size in USD, lock duration, whether LP tokens have been burned. The difference between a legitimate launch and a rug-in-waiting is almost always visible here.
-
low_liquidityTier-based by USD depth-5 to -30
-
liquidity_not_lockedCan be pulled any block-15
-
liquidity_not_burnedSignal, not deal-breaker-5
Who already owns this?
Top-10 holder concentration, deployer's own position, and a real coordinated-bundling check. Five minutes after detection we pull the first ~15 buys and score them on timing density, amount uniformity, and wallet diversity. Coordinated launches don't slip past as 'organic'.
-
high_top10_concentration30% / 50% / 70% tiers-10 to -30
-
high_deployer_holdings5% / 15% / 30% tiers-10 to -30
-
bundling_detectedReal first-buy analysis (live)-25
Real activity or machine noise?
Wash-trading detection, organic holder growth, legitimate trade patterns. A token with a hundred wallets trading with itself is a different beast than one with a hundred distinct buyers.
-
wash_trading_detectedSelf-trade pattern match-20
-
low_holder_count50 / 200 holder tiers-5 to -15
Did they even try?
Website presence, social media footprint. Small signals, but they stack up. A project with zero internet presence at launch is rarely the one you want to be early on.
-
no_websiteExpected baseline-5
-
no_social_mediaOrganic reach check-5
Has this wallet shipped rugs before?
Every token gets cross-referenced against the deployer's full history. Reputation is a 0–100 score aggregated from every prior token TokenRadar has scored. Serial ruggers (≥5 rejected, reputation ≤30) get a hard −30 flag; clean records (≥3 tokens, reputation ≥80) surface as a positive cue. Public leaderboard at /deployers.
-
deployer_serial_ruggerPattern of rejected tokens-30
-
deployer_clean_recordConfidence cue, no penaltyinfo
When does the LP get pulled?
We parse Streamflow LP locks straight off Solana. Token cards show a live countdown to unlock; the score is downgraded as the unlock window approaches, and watchlist users get a heads-up before LP becomes withdrawable. Already-unlocked LPs trip a separate flag.
-
liquidity_unlocks_soonTiered: <1h / <24h / <1w-5 to -30
-
liquidity_unlocked_alreadyLock has expired-30
Route alerts to where you already live.
Configure per-profile delivery. A tight sniper profile hits your Telegram in under a second. A research profile goes to your panel with a daily digest.
Live feed in the web dashboard with sub-second updates and inline labels.
Rich messages with token card, score, flags, and a deeplink to your trade bot of choice.
HTML alerts with classification badge — plus daily digests for slow-moving research profiles.
Your endpoint, our payload. Pick from seven bot-ready templates or use generic JSON.
Pick a bot.
We craft the deeplink.
Choose a format from the dropdown when you build a profile. Your alerts arrive shaped exactly how the receiving bot expects them — no middleware, no copy-paste of token addresses.
See your first flag inside five minutes.
Register, build a profile, start scanning. No setup wizards. Nothing to install.