How to Read a DeFade Scan

Every scan returns one headline score and a stack of modules underneath it. This page explains what the score bands mean, what each module is actually answering, and how to read it without guessing. No jargon, no maths homework.

Scan a token →
The score Token header Overview Whales Intel Trading Smart Money Behind the score Badges & labels Beyond the scan
The rug pull score

One number from 0 to 100. It counts up toward danger — 0 is clean, 100 is as bad as it gets. It is a weighted composite of the modules below, not a vote: a single disqualifying finding can set a floor the rest of the report cannot average away. The colour of the score is the fastest read on the page.

0–24
PROBABLY SAFE
Minimal risk indicators. Liquidity and holder distribution look healthy and no major flag fired. Low risk, not zero risk.
25–44
SOME RISK
A few risk factors present. Worth opening the modules that fired before you decide anything.
45–59
ELEVATED RISK
Several signals stacking up. This is the band at which a token is automatically recorded to the rug database.
60–74
HIGH RISK
Significant signals: bundled wallets, concentrated holdings or suspicious dev activity.
75–100
EXTREME RISK
Strong rug pull indicators. Tokens that have already collapsed skip the bands entirely and come back as RUGGED / DEAD or ABANDONED / DEAD.
The score is never shown half-finished. If a check is still running, the scan holds the number back rather than flashing an optimistic one that moves later. And where a source cannot be reached, the affected module says what it could not see instead of guessing — an unverifiable holder is never counted as safe.
Reading it in 30 seconds
  1. Read the colour first. Green and yellow mean "keep reading". Orange and red mean the burden of proof has flipped — you now need a reason to buy, not a reason to walk.
  2. Check the dev wallet. A creator who has already sold or moved supply out on a token that is hours old is the single loudest signal on the page.
  3. Check the launch. Heavy bundled supply plus snipers connected to the deployer means the exit was planned before you arrived.
  4. Check who the holders really are. Fifty wallets funded from one source is one holder, not fifty. The funding and clustering modules are what separate the two.
  5. Check the exit. A great-looking chart with thin liquidity is a position you cannot leave. The exit liquidity calculator answers that directly.
Overview Tab 1

The shape of the token: who holds it, what the liquidity looks like, who launched it and how long it has been alive. If you only read one tab, read this one.

Rug Pull Probability FREE

What it answers
How likely is this token to be a rug, on a single 0–100 scale?
How to read it
The card lists the signals that actually contributed, so you can see why the number is what it is. A high score built on one flag reads differently from the same score built on five.

Liquidity Analysis FREE

What it answers
How deep is the pool, is the LP burned or locked, and how does liquidity compare to the market cap it is supporting?
How to read it
Burned or locked LP removes the one-click rug. Unlocked LP that is tiny next to the market cap is the classic setup — the chart can look enormous while the actual pool is a few thousand dollars deep.

Holder Distribution FREE

What it answers
How much of the supply sits with the top holders, once pools, burns, lockers and exchange wallets are taken out of the count?
How to read it
The filtering is the point — raw top-holder lists usually put the liquidity pool at number one and look terrifying for no reason. Read the filtered concentration: if a handful of real wallets can collapse the chart, they eventually will.

Token Lifecycle FREE

What it answers
What has happened to this token since it was created, on a timeline.
How to read it
Age is context for everything else. A concentrated holder base on a two-year-old token is normal; the same picture two hours after launch is not.

Deployer Reputation PRO

What it answers
What happened to the other tokens this deployer has launched?
How to read it
Survival rate is the number that matters. A deployer whose previous launches all died within days is telling you what this one is for. The previous tokens are clickable, so you can check the pattern yourself.

Exit Liquidity Calculator FREE

What it answers
If you sold a given position right now, what would it actually do to the price — and what is the largest size you could exit cleanly?
How to read it
Type in the size you are actually considering. If the slippage on your own position is ugly, the exit is the risk, regardless of what the rest of the report says.

Social Sentiment Timeline PRO

What it answers
How much is this token being talked about, how that has moved over the last week, and how the DeFade community has voted on it.
How to read it
Direction beats level. A mention spike with nothing on-chain behind it is a promotion, not demand. Sentiment never feeds the rug score — it is context, not evidence.
Whales Tab 2

Who the big holders are, where their money came from, and which of them are really the same person. This is the tab that separates "many holders" from "one holder in many costumes".

Whale Alert PRO

What it answers
Which individual wallets are big enough to move the price on their own, and what share each one controls.
How to read it
Ask what happens if the top one leaves. If a single wallet's exit would erase the chart, that wallet is your real counterparty.

Bubble Map FREE

What it answers
The holder base as a picture: one bubble per wallet, sized by holding, with lines where wallets are connected.
How to read it
Look for shape, not detail. Scattered unconnected bubbles are a real crowd. A tight web of linked bubbles holding most of the supply is one group.

🔗Funding Origin PRO

What it answers
Where the money that bought this token came from, traced back to its source and grouped by how much identity checking that source does.
How to read it
A holder base funded through mainstream, identity-checked venues behaves differently from one funded through mixers, bridges or no-KYC routes. Mixer and bridge inflows are called out explicitly.

🕸️Cross-Funding Graph PRO

What it answers
Did the top holders fund each other before the launch?
How to read it
Each line is one wallet sending to another. Independent buyers do not have lines between them. A connected chain among the largest holders means the "crowd" was assembled, not gathered.

👥Sybil Band Clustering PRO

What it answers
Collapses those funding links into bands — groups of wallets that behave as one actor — and tells you how much supply each band really controls.
How to read it
Read the band's combined percentage, not the individual wallets. Shared exchange, mixer and bridge sources are excluded, so wallets are never grouped just for having used the same exchange.

Fee Fingerprint PRO

What it answers
Which wallets were driven by the same trading bot. Every terminal lets an operator pin a priority fee and a Jito tip, and those settings ride along in every transaction the bot signs — so wallets paying the exact same compute-unit price were probably running one person's configuration. It catches the operator who sold their launch wallets and re-entered through fresh ones: new wallets, same settings. Solana only.
How to read it
This is a linkage signal, not a bundle — it says nothing about timing, and two wallets can match days apart. Read the confidence figure on each group and the note beneath it. A round number like 1000 is something strangers pick independently; an odd one that keeps reappearing across separate sessions is the strong case. Values that most of the launch used are excluded as bot defaults rather than reported, and the card tells you how many wallets that covered.

Top Holder History PRO

What it answers
How the top holders' positions have changed over time, with each whale labelled accumulating, distributing, holding or new.
How to read it
Quiet distribution shows up here before it shows up on the price chart. A rising price with the top holders labelled "distributing" means you are the exit.

Terminal PRO

What it answers
A live running feed of the token's notable on-chain activity as it happens.
How to read it
Use it while you are actually in a position. Everything else on the page is a snapshot; this is the one view that keeps moving.
Intel Tab 3

The launch and the people behind it: who the creator is, what they have done with their own supply, and whether the first buyers were insiders.

Dev Wallet Tracker FREE

What it answers
What the creator has done with their own supply on this token: still holding, sold, or quietly moved it elsewhere.
How to read it
This is the loudest single card on the page. Fully exited or transferred out on a young token is a walk-away. Transferring out is worse than selling — it hides the supply rather than removing it.

Creator Intelligence FREE

What it answers
Who the creator is as a wallet: how old it is, and whether it launches tokens as a habit.
How to read it
A brand-new wallet launching its first token has no track record to defend. A serial deployer has one — go and read it in Deployer Reputation.

Bundle Detection Pro PRO

What it answers
Did a coordinated group buy in the launch window, and how much of the supply did they take?
How to read it
Bundled supply is the strongest single predictor of a coordinated dump. Each group carries a confidence level and shows what those wallets have done since — bought and still holding is a loaded gun; bought and already selling is the trigger being pulled.

Historical Bundles PRO

What it answers
The same question, but for the pumps that happened after launch rather than at it.
How to read it
A clean launch does not mean a clean token. Coordinated buying during a later volume spike is a manufactured pump — and the wallets behind it are usually still holding when it unwinds.

Fresh Wallet Detection PRO

What it answers
How many of the top holders are wallets that were created right before the token was, and how much supply they hold between them.
How to read it
The flag is tiered by how fresh the wallet is relative to the launch — the tighter the gap, the louder the signal. Real buyers usually have a history. Wallets born hours before the token normally belong to the launch, not to the market.

Holder Network PRO

What it answers
The insider map: transfers and shared funders between the top holders, drawn as a graph.
How to read it
Clusters are the finding. A dense cluster wired back to the deployer is a dev-linked group hiding behind separate addresses.
Trading Tab 4

How the token is being traded and by whom — the bots at the front, the wallets copying each other, and the accounts you might recognise.

Token Snipers PRO

What it answers
Which wallets got in within seconds of the launch, what they spent, and what they still hold.
How to read it
Snipers alone are normal on any launch. Snipers connected to the deployer's funding are not — that combination is a pre-arranged position, not a race. Watch the holding / sold / transferred status per wallet.

Copy Trading Detection FREE

What it answers
Which wallets are mirroring each other's trades on this token.
How to read it
Mirroring cuts both ways: it can be a copy-trade crowd following one good wallet, or one operator running many. Cross-check the same wallets against the funding and clustering modules to tell which.

🌐Holder Overlap PRO

What it answers
What else do this token's top holders hold? Compare up to three tokens at once and see which other tokens their wallets share.
How to read it
Shared tokens appear as bubbles. A heavy overlap with one other token means the same group is running both — useful for spotting a narrative rotation, and for spotting the same operators recycling.

KOL Holders FREE

What it answers
Which known influencer wallets are currently holding the token, and how much.
How to read it
Informational only — it never moves the rug score. A KOL holding is not an endorsement, and a KOL holding a lot before a call is the setup people usually find out about too late.

Wallet Profiler PRO

What it answers
Paste any wallet from the report and get its trading record: win rate, realised profit and loss, and how long it typically holds.
How to read it
Use it on the top holder that worries you. A wallet with a long profitable record buying early is a different story from a wallet with no record at all.
Smart Money Tab 5

The only tab that can be positive. Everything else looks for reasons to walk away; this one looks for wallets with a record of getting it right.

Smart Money Tracker PRO

What it answers
What share of this token's holders are wallets with a genuine track record, versus ordinary retail?
How to read it
A high smart-money share is a point in the token's favour — but it is never a safety verdict. Experienced wallets trade rugs on purpose all the time; they just leave first.

Smart Money Signals PRO

What it answers
Are those wallets buying right now?
How to read it
Live activity rather than a standing position. Read it alongside Top Holder History — smart wallets buying while the top holders distribute is a contradiction worth resolving before you act.
Behind the score

Three modules that do not have a card of their own. They work across the whole report and show up as flags inside the modules above.

Signal Fusion PRO

What it answers
Which wallets were caught by more than one module independently.
How to read it
A wallet flagged as a sniper and a fresh wallet and an insider is one operator wearing three disguises — each module on its own only sees one of them. Wallets that appear in several modules deserve the most attention.

Wallet Reputation PRO

What it answers
Has this wallet turned up on tokens that later died?
How to read it
A note like "seen on 4 previously rugged tokens" is a repeat-offender flag. It is the one signal that fires before the wallet does anything on this token, and it gets sharper the longer the platform runs.

Contract Safety FREE

What it answers
On EVM chains: is the contract a real token, and has ownership been renounced?
How to read it
An owner who has not renounced may still hold powers over the token. This is the EVM counterpart to Solana's mint and freeze check, and it feeds the score the same way.
Badges & labels

Small things that appear all over the report and mean the same thing everywhere.

PROPro module

The module exists on every scan but its detail is unlocked on a paid plan. On the free tier you see the card and the lock rather than a missing section — so you always know what was not checked. The first scan of each day runs the complete analysis for everyone. Plans.

Could not verify

When a module cannot see something, it says so instead of assuming the best. Treat "could not verify" as unknown, never as clean — an unverifiable holder is not a safe holder.

Chain coverage

Solana, Ethereum, Base and Robinhood Chain, detected automatically from the address. Every module is rebuilt against each chain's own mechanics rather than relabelled — a handful are Solana-specific, and the report tells you which ones did not apply.

What this is not

DeFade quantifies known risk patterns. A low score means the common rug setups are absent — that is risk reduction, not a promise. Memecoins are high-risk speculative assets and nothing here is financial advice.
Beyond a single scan

Each scan answers "is this token safe". These pages answer the questions that span many tokens — and they are all in the menu, top left.

Rug Feed

Every token the engine has flagged, newest first, searchable per chain.

Survival

How long scanned tokens actually live after launch, and where a given token sits on that curve.

Deployer Lookup

Paste a deployer, or any token it launched, and read its full launch record across every chain.

Wallet Lookup

Any address's cross-token record: what it was flagged as, on what, and which of those tokens later died.

KOL Leaderboard

Which callers called what, and how those calls actually aged.

Whitepaper

The full module list and the scoring methodology in more depth, plus the roadmap.
Still stuck on a number? The FAQ answers the common ones, 10 red flags walks through a real token signal by signal (with more write-ups on the blog), and the API docs document the exact response shape if you are consuming a scan programmatically.