Eleven checks run against every tokenized stock in the universe: how far its
on-chain price sits from a reference price, how much liquidity stands behind it, whether the
24-hour flow looks like traders or bots, what share of the sampled swaps reverted, how
concentrated the holders are, how strong the issuer's reserve evidence is, how the authority
keys are held, whether trading is paused, whether any top-20 account is frozen, and how far
apart the venues price it, plus whether its tech and legal template lets a permissionless
smart-contract lender hold collateral and enforce a default. Three of the eleven — the legal
evidence, the authority keys and DeFi enforceability — describe the issuer's
programme and give every token of that issuer the same answer, so each token gets two
verdicts: its programme's, and its own from the other eight, with the number of those checks it
passes. A token is only called good on its own checks if at least one of its market checks
could be run.
A check whose inputs are missing is unknown, and an unknown is never
counted as a pass or as a fail.
Loading…
Token addresses this pipeline first saw in the last 14 days. “First seen” is when Jupiter first returned the address to us, which can be later than the token's creation. The first eight are shown below; open the rest only if you need them.
The tiles count each token on its own eight checks: price tracking, liquidity, organic flow, failed swaps, holder concentration, venue spread, trading pause and frozen accounts. Each tile filters the table below; click it again to clear, and click a second status to see both. Picking a status does not change the tiles' own counts: with warning selected, the other three still show how many tokens you would get by switching to them.
Programme is the worst of the three issuer-wide checks: legal evidence and reserve verification, how the authority keys are held, and whether a DeFi lender could enforce the token as collateral. Every token of an issuer gets the same answer. Worst of all eleven is the single badge this page used to lead with; it is kept so nothing is hidden, but it cannot tell two tokens of one issuer apart.
The same eleven checks grouped by what they are about: market quality, on-chain control, the evidence that the off-chain shares exist, and whether a DeFi protocol can enforce custody and default. “Unknown” means we did not measure it.
Each bar counts the tokens whose own checks fail, under the first check that does.
The panel lists every property you can filter by, with the count of tokens behind each value. Two values inside one group are OR; two groups are AND, and each group's counts leave out its own selection, so a count always says what you would get by clicking it. A symbol links to that token's card. Premium is the on-chain price against the reference price for the same underlying share; spread is the gap between the cheapest and dearest venue pricing it; When closed names each lending market that takes the token and the price it uses while the US market is closed: frozen at close, 24/5 overnight, 24/7 token price, signed quote or stale since a date (hover for the liquidation threshold; the token's card has the freezes, Monday gaps and depth; the closed-market data is downloadable). Last trade is the newer of our own trade tape and CoinGecko's exchange ticker, and says which; the ticker is refreshed for a few hundred tokens a day, so it can lag. A dash is a missing measurement, never a zero.
| Token | Issuer | This token | Programme | Market | Control | Legal / evidence | DeFi composability | Liquidity | Premium | Venue spread | When closed | Top-1 share | Last trade |
|---|
Only the columns with an arrow are sortable. A sort runs over every matching token, not just this page. This token sorts good, caution, warning, then by how many of its checks fail and how many pass; a token with no market sorts last either way. Last trade sorts by the CoinGecko time.
A snapshot of every tracked token is saved every day and consecutive days are compared. Only these changes are reported: a pause, a rebase, a pool that halved from real liquidity, a venue spread crossing 5 %, a health status changing, a frozen account appearing, a control flag flipping. A field that was not measured on either day cannot produce a change. The first snapshot may therefore contain many newly measurable fields without implying that all of them changed on-chain that day.
A separate daily snapshot asks whether each exact token address still appears in each protocol's checked registry, whether maximum LTV changed, whether a live market became inactive, and whether reported lending collateral fell sharply. “Removed” means no longer observed in that registry. Other routes into the protocol may still work. Collateral alerts require both a fall of at least 25% and a previous value of at least $100,000; missing measurements never become zero.
Every row retains the exact Solana token address and the two observation times. The same evidence is downloadable as DeFi changes data.
Hand-picked, dated incidents from the issuer dossiers, each recorded with its source. The list is curated by hand and is not updated automatically.
One row per Meteora pool, with the facts DexScreener does not report: which Meteora product the pool is (DexScreener files the DLMM pools and the bonding curve under one name), its bin step, its real fee configuration and its 24-hour fees as well as volume. Premium is that pool's own price against the token's reference price, so a pool quoting away from the venue aggregate shows up. Failed is the reverted share of the signatures the trade collector sampled on that pool. A dash means the collector never reached the pool.
| Token | Pool | Bin step | Fee | Liquidity | 24 h volume | 24 h fees | Premium | Failed / last trade |
|---|