Seven numbers. Each one is backed by transactions in the sections that follow and by the files in the data folder.
Report 01 was built from data as of 2026-10-03 07:29 UTC. Its figures stay as of that time. This edition adds results that did not exist then.
| Measure | Oct 3 7:29 Report 01 data | Oct 5 12:03 daily run | Oct 5 12:13 daily run |
|---|---|---|---|
| Launches indexed | 664,689 | 673,621 | 673,649 |
| Deployer addresses | 386,627 | 392,413 | 392,428 |
| Launches called rugged | 3,708 | 3,979 | 3,979 |
| ETH taken in those rugs (net) | 18,268 | 18,839 | 18,839 |
| Proven bundled launches | 115,919 | 135,312 | 135,457 |
| ETH taken by bundles (net) | 44,305 | 44,867 | 44,851 |
| X posts indexed | 49,670 | 71,469 | 71,469 |
| Proven wash share of volume, % | 1.3 | 1.3 | 1.3 |
| Sniper wallets | 1,771 | 1,771 | 1,771 |
Between the first and last column 52.7 hours passed and 8,960 launches were indexed. A launch counts as rugged or bundled only when its prices and sells have been read, so a young launch can move between columns; the daily run records the re-judged count separately.
Detectors ran over the 30,014 coins launched Sep 21 to Oct 5 with 10 or more buyers. A coin gets a type only when its rule holds and a transaction proves it. Every hit is in the files rug_hits.csv with its proof link.
| Type | Coins | ETH | Median time | Not red | On-chain check | Proof |
|---|---|---|---|---|---|---|
| Creator dumps | 782 | 10,823 ETH net profit 9,990 ETH | 88 seconds | 438 56.0% | 20 of 20 | $ROB $VRAX $FOOMS |
| Liquidity pulls | 4,830 | 31,198 ETH | 11 minutes | 4,717 97.7% | 20 of 20 | $SWAP $ZERO $AICABAL |
| Operation self-snipes | 179 | 93.7 ETH | under a second | 44 24.6% | 20 of 20 | $SI $DOT $MSAI |
| Serial actors | 748 | 44 operations 19 creator wallets | n/a | 284 38.0% | 20 of 20 on each of the three rules | see tables |
| Coordinated exits | 179 | 782 ETH | 22 minutes | 120 67.0% | 20 of 20 on a second sample | $HOOD6900 $SI $PKRT |
| Insider clusters | 84 | 339 ETH | 29 minutes | 64 76.2% | 20 of 20 on a second sample | $HOOD6900 $SI $PKRT |
| Wash then dump | 34 | 628 ETH | 50 seconds | 4 11.8% | 20 of 20 | $DEED $APES $APES |
A tie is an on-chain fact: the anchor wallet funded the wallets, received their ETH, airdropped to them, or passed the coin to or from them. Busy bot wallets and hubs never anchor a group. ETH figures count only ETH or WETH quotes and drop any single figure of 10,000 ETH or more as a price artefact. Every figure is a floor: exits through wallets the creator funded but did not sign from are not counted.
Creator dumps by launch venue: 624 coins launched through Pons, 156 through Uniswap v3, 2 through Uniswap v2 and v4. Liquidity pulls are counted on Uniswap v2 (3,587) and v3 (1,243) only. Pons is infrastructure and is never the subject: the coins are, and the wallets that launched them.
By day of launch. The first and last day are partial. Green or unread coins are the ones the new detectors add.
A coin the launch detector read as green, explained from the first funding transaction to the cash-out.
$ASTRAL launched on Pons on 2026-10-03 20:13:58 UTC from a throwaway creator wallet. The launch detector read it as clean because the buying wallets share no funder except trading-bot relayers. The wallets that bought it in the first second belong to one operation: 62 launches by 46 creator wallets, 38 fee wallets and 22 collector wallets, linked by creator-fee money. The operation took 4.87 ETH from this coin, and its snipe wallets were out within 12 minutes. The 98% price collapse came later and was a crowd sell-off.
| Coin | Launched | Wallets | First-minute buy | Profit | Proof |
|---|---|---|---|---|---|
| $Sesbot | Sep 22 | 1 | 12.3% | 0.948 ETH | proof tx |
| $geth | Sep 23 | 4 | 12.8% | 0.418 ETH | proof tx |
| $AZ | Sep 23 | 5 | 10.6% | 0.909 ETH | proof tx |
| $CODOS | Sep 26 | 4 | 17.5% | 0.268 ETH | proof tx |
| $GLMS | Sep 29 | 6 | 11.0% | 0.265 ETH | proof tx |
| $ASTRAL | Oct 3 | 10 | 19.0% | 0.532 ETH | proof tx |
| Total | 3.34 ETH |
The 62 launches of the operation are listed in astral_operation_launches.csv. The case file by agent rugdeep, with every transaction re-read on the public RPC, is in cases/astral. Operation sizes differ by definition: the fee-money graph reaches 62 launches in 28 days, while the case file counts the 35 Pons launches whose fee wallets forward to the one collector 0x3a16 (40.08 ETH claimed in total).
19 creator wallets have a proven dump or pull on 3 or more coins (144 coins). 44 fee-money operations launched coins with a proven exit on 2 or more other coins (582 coins). The tables rank the largest by ETH, each row with its proof.
| Actor | Coins | ETH taken | Active | Proof |
|---|---|---|---|---|
| 0x49de94…18d9 creator wallet, Uniswap v3, 2 of its coins reuse an earlier ticker | 3 | 102 | Oct 3 to Oct 3 | largest: 55.1 ETH $CLAUDIA, creator dump |
| 0x517d9e…71a8 creator wallet, Pons, 5 of its coins reuse an earlier ticker | 25 | 33.3 | Sep 22 to Sep 25 | largest: 2.65 ETH $SGI, creator dump |
| 0x25cab0…6dd6 creator wallet, Pons | 16 | 30.0 | Sep 22 to Sep 24 | largest: 2.69 ETH $STONKYARD, creator dump |
| 0xb62af8…a6bb creator wallet, Pons, 2 of its coins reuse an earlier ticker | 14 | 23.6 | Sep 21 to Sep 22 | largest: 2.15 ETH $AGE, creator dump |
| 0xc4ff18…cd7c creator wallet, Uniswap v3 | 3 | 21.6 | Sep 24 to Sep 24 | largest: 7.50 ETH $Nautilo, liquidity pull |
| 0x01f72d…f200 creator wallet, Pons, 4 of its coins reuse an earlier ticker | 21 | 18.2 | Oct 1 to Oct 3 | largest: 1.34 ETH a coin, creator dump |
| 0x707da7…8ef4 creator wallet, Uniswap v3 | 3 | 7.10 | Sep 28 to Sep 28 | largest: 6.16 ETH $boar, creator dump |
| 0xb3f612…6f55 creator wallet, Pons | 3 | 5.32 | Sep 23 to Sep 23 | largest: 2.08 ETH $ETHP, creator dump |
Also re-read on the public RPC: the proof transactions of the 3 largest creator wallets: 3 of 3 succeeded and were signed by the wallet. A coin reuses an earlier ticker when its symbol equals that of an earlier coin with 2,000 or more buyers.
ETH taken is the sum, per coin, of the larger of the creator's net sale profit and the ETH removed from the pool above the creator's own deposit. A floor: a coin counts once and exits through wallets the creator funded but did not sign from are not counted. Window: coins launched in the last 14 days.
| Threat actor (fee-money operation) | Exit coins | Exits, ETH | Fees, ETH | Total, ETH | Proof |
|---|---|---|---|---|---|
| Operation of 62 launches 0x3a163e…ba62 46 creator wallets, 38 fee wallets, 22 collectors | 6 | 3.34 | 36.8 | 40.1 | largest exit 0.95 ETH largest claim 8.23 ETH |
| Operation of 78 launches 0x275940…da9f 76 creator wallets, 76 fee wallets, 81 collectors | 29 | 20.0 | 13.6 | 33.6 | largest exit 1.25 ETH largest claim 0.42 ETH |
| Operation of 7 launches 0xd00e03…c005 7 creator wallets, 7 fee wallets, 8 collectors | 3 | 4.72 | 22.1 | 26.8 | largest exit 3.82 ETH largest claim 9.84 ETH |
| Operation of 6 launches 0x0f2528…daf7 6 creator wallets, 6 fee wallets, 10 collectors | 6 | 11.9 | 14.4 | 26.3 | largest exit 8.49 ETH largest claim 4.53 ETH |
| Operation of 30 launches 0xc8fb11…50f3 30 creator wallets, 30 fee wallets, 26 collectors | 11 | 11.1 | 0.97 | 12.1 | largest exit 2.29 ETH largest claim 0.07 ETH |
| Operation of 67 launches 0x51a9b4…ed87 67 creator wallets, 67 fee wallets, 182 collectors | 5 | 4.87 | 6.69 | 11.6 | largest exit 1.45 ETH largest claim 0.42 ETH |
| Operation of 82 launches 0xe6578d…e14a 76 creator wallets, 76 fee wallets, 5 collectors | 14 | 4.47 | 3.93 | 8.40 | largest exit 0.47 ETH largest claim 0.28 ETH |
| Operation of 10 launches 0x11e6f1…a81b 8 creator wallets, 8 fee wallets, 1 collectors | 3 | 1.51 | 2.13 | 3.63 | largest exit 0.90 ETH largest claim 0.47 ETH |
An operation is a set of fee wallets that forward their claimed creator fees to the same collector wallets within 72 hours. Exits are the per-coin largest proven take of the operation's coins (creator dump profit, pool removal above deposit, or self-snipe profit). Creator fees are the claims of its fee wallets in the last 28 days and are shown apart because Pons pays them to creators by design; the combined column ranks the table. The operation of $ASTRAL leads.
Window: 2026-10-02 07:14:00 UTC to 2026-10-05 12:23:51 UTC, 77.2 hours, up from 28.8 hours in Report 01. About two days of Solana launchpad activity measured from the first slot our own index holds. It is not the history of Solana.
| Launchpad | Analysed | Rug operations | Share | SOL taken, net | Deployers | First sell |
|---|---|---|---|---|---|---|
| pump.fun | 161,332 | 9,856 3,745 in Report 01 | 6.1% | 175,658 64,647 in Report 01 | 5,548 | 2 s |
| Raydium LaunchLab | 10,159 | 3,056 1,576 in Report 01 | 30.1% | 32,086 17,906 in Report 01 | 519 | 3 s |
| Meteora DBC | 17,567 | 58 29 in Report 01 | 0.3% | 1,048 495 in Report 01 | 53 | 13 s |
| All | 189,058 | 12,970 5,350 in Report 01 | 6.9% | 208,792 83,048 in Report 01 | 5,856 | 3 s |
A rug operation is a red launch where the deployer-linked cluster sold at least 80% of what it bought and made at least 0.5 SOL, or pulled liquidity. Money is counted from these rows only. Bundled launches (9,298) are a count and are never netted. Takes are the pool-side amount, gross of network fees; a launch young at the time of the read can still grow. Report 01 figures are those frozen at its build.
| Token | Launchpad | Deployer | SOL taken | Bought / sold | USD | Proof |
|---|---|---|---|---|---|---|
| MrBeast qptxT4..pump | pump.fun | 9zcotT..wBcy | 4,692 | 331 / 5,024 | $562K | launch tx |
| MEMEAGENCY CyChQH..pump | pump.fun | CftPeL..sH9q | 3,505 | 140 / 3,645 | $425K | launch tx |
| PUMPGO 7gbJ4S..pump | pump.fun | 5mvjTR..WfTs | 2,976 | 234 / 3,210 | $351K | launch tx |
| SUKI Ay9V3r..pump | pump.fun | 5ngMxW..FeZD | 2,717 | 205 / 2,921 | $324K | launch tx |
| ANON 2AKrTT..pump | pump.fun | 4SxAN2..2Vin | 2,373 | 215 / 2,588 | $285K | launch tx |
| DOTF 42qptE..pump | pump.fun | 8RYP9G..fnJt | 2,275 | 4,046 / 6,320 | $278K | launch tx |
| SUKI AW7xdc..pump | pump.fun | 7xpY9v..33NW | 2,212 | 215 / 2,426 | $264K | launch tx |
| DOTF 3hr14K..pump | pump.fun | 7WFwgB..fNL8 | 2,145 | 4,046 / 6,190 | $260K | launch tx |
Size of the takes: 1,491 under 1 SOL, 8,170 from 1 to 10, 3,058 from 10 to 100, 232 from 100 to 1,000 and 19 of 1,000 SOL or more. The median is 4.99 SOL. 662 of 5,856 deployer wallets have 2 or more rug operations (291 with 5 or more, 180 with 10 or more).
| Deployer wallet | Rug operations | SOL taken | Launchpads | Active |
|---|---|---|---|---|
| 8Wz4CQpT..ahYWF | 173 | 866 | pump.fun, Raydium LaunchLab | Oct 2 to Oct 5 |
| E4FPggw9..LpZEA | 150 | 2,112 | pump.fun, Raydium LaunchLab | Oct 2 to Oct 5 |
| 29yFzeBZ..ikbw8 | 133 | 849 | pump.fun, Raydium LaunchLab | Oct 2 to Oct 5 |
| GpTXmkdv..8WAc4 | 119 | 1,165 | pump.fun, Raydium LaunchLab | Oct 2 to Oct 5 |
| GeUnv1jm..aUX1W | 111 | 363 | pump.fun, Raydium LaunchLab | Oct 3 to Oct 5 |
Oct 2 starts at 07:14 UTC and Oct 4 is partial.
Four teams added 17 detections for what the first seven rug types miss, and 2 more for Solana. Each went live on the Fable cards only after an independent re-check of random hits, on chain or in a simulation: 19 of 20 or better, or every hit when fewer than 20 exist. Every row has transactions or posts you can open.
With the four wider rug detections added (coordinated exits, insider clusters, wash-then-dump fleets and small liquidity pulls), 6,711 coins (22.4% of 30,014) carry a proven rug type, up from 5,867 with the first seven detectors alone. The wider rules overlap the first seven.
| Detection and rule | Hits | Coins | Independent check | Proof |
|---|---|---|---|---|
| Coordinated exit 3 or more tied wallets sold 10% or more of supply inside their densest 15 minutes, the price was 50% or more lower in the next hour and outside buyers were net 0.5 ETH or more in. Any entry time, no price run needed. | 679 | 679 | 19 of 20 on chain, 0 false reds | tx 1 tx 2 tx 3 |
| Insider cluster The coordinated-exit rule where the group's first buy came after the launch minute and it bought 10% or more of supply. | 120 | 120 | 19 of 20 on chain, 0 false reds | tx 1 tx 2 tx 3 |
| Wash then dump, tied fleet 5 or more tied wallets bought 20% or more of supply and sold 90% or more of it with a profit or loss within 20% of cost, made 30% or more of all volume, started before the price first halved, on a coin now at 30% of its peak or less. | 900 | 900 | 20 of 20 on chain | tx 1 tx 2 tx 3 |
| Small liquidity pull One removal transaction took 90% or more of a Uniswap v2 or v3 pool when only 0.05 to 0.5 ETH was left in it, on a coin that is not red. Amber, never red. | 25 | 25 | 20 of 20 on chain (the first rule scored 15 of 20 and was replaced) | tx 1 tx 2 tx 3 |
| Sells blocked (honeypot) On a Uniswap v2 or v3 coin with live liquidity, real holders each try to sell 1% of their balance through the router in a simulation and every sell reverts, a fresh wallet can buy but cannot sell, and 3 or more sells were refused on chain. | 41 | 41 | 19 of 20 by simulation, a fresh wallet 20 of 20 | tx 1 tx 2 tx 3 |
| Sell tax of 10% or more A simulated router sell by a real holder works, but 10% or more of the coins leaving the holder never reach the pool. A plain transfer into the pool and a fresh wallet's sell show the same cut. | 9 | 9 | 9 of 9 by simulation | tx 1 tx 2 tx 3 |
| Owner powers proven by simulation An owner function (blacklist, trading switch, fee, mint or sale cap) is matched to the coin's code, then called by the owner in a simulation, and a holder's next sell reverts or is taxed 50% or more. Ownership is not renounced. | 18 | 18 | 19 of 19 by simulation | tx 1 tx 2 tx 3 |
| Hidden concentration Among the top 20 holders, 3 or more wallets tied by a funder, a shared sweep wallet or the coin passed between them hold 10% or more of supply together and 5% or more beyond the largest wallet. A fact about the holder list now. | 94 | 94 | 20 of 20 on chain, 0 false reds | tx 1 tx 2 tx 3 |
| Fake holders One sender moves a coin to 20 or more wallets in one transaction with no swap. Wallets that hold exactly that drop and nothing else are 100 or more and 30% or more of the current holders. Amber from 50% of holders, otherwise information. | 118 | 118 | 20 of 20 on chain (611,845 transfer logs replayed) | tx 1 tx 2 tx 3 |
| Thin liquidity Past the Pons curve, a coin with a market cap of 20,000 USD or more where all priced pools would pay out under one fiftieth of it if sold to zero, and a 1% sell moves the price 25% or more. | 25 | 25 | 20 of 20 under the card wording (first draw 17 of 20) | tx 1 tx 2 tx 3 |
| Price painting Drip: one wallet, or the wallets of one funder, make 30 or more buys, 20% or more of the coin's buys, with a median size of 0.002 ETH or less. Loop: 10 or more buys and sells, flat position, turnover of 10 or more, 1 ETH or more and 20% or more of the volume. | 96 | 96 | 19 of 20 on chain | tx 1 tx 2 tx 3 |
| Front-running A wallet buys at most 6 blocks (1.5 seconds) before a buy of 0.3 ETH or more and sells 60% to 150% of it within a second after, at a profit, 5 or more times, on a coin with 3 or more victims. | 375 | 141 | 20 of 20 on chain (the first rule scored 16 of 20 and was replaced) | tx 1 tx 2 tx 3 |
| Unlinked snipers Wallets not tied to the creator or the launch bundle each buy 0.3% or more of supply in the first 3 seconds, sell 80% or more within 5 minutes at a profit, and together held 5% or more of supply. | 207 | 207 | 19 of 20 on chain | tx 1 tx 2 tx 3 |
| Coins named in phishing posts Robinhood coins whose contract is named in posts that invite a listing vote or a token claim and link to a look-alike domain. Information only: the coin is the bait, not the rug. | 242 | 224 | 20 of 20 on live posts | post 1 post 2 post 3 |
| Phishing accounts on X Accounts whose posts invite a listing vote or a token claim and link to a short link or look-alike domain that 3 or more accounts used. The link is not the named brand's site. Accounts are counted, not named. | 92 | 92 | 20 of 20 on live posts | post 1 post 2 post 3 |
| Same-text shill campaign, coin 3 or more accounts post near-identical text with the same contract within 30 minutes. Tool accounts, list posts, warnings and bait templates are excluded. Information line. | 2 | 2 | 1 of 1 on live posts (every hit) | post 1 post 2 |
| Shill networks on X An account that is in 2 or more same-text campaigns on different coins, with 2 or more of the same partner accounts. Accounts are counted, not named. | 13 | 13 | 10 of 10 on live posts | post 1 post 2 post 3 |
Hits are the rows the cards serve at build time. Coins are distinct coins, or accounts for the X kinds. The coordinated-exit, insider-cluster, wash-then-dump and liquidity-pull rows are recomputed hourly over the last 14 days; the honeypot, hidden-concentration, thin-liquidity and fake-holder rows are the coins read in the latest refresh. Proof links are three examples of the latest hits.
Our review set holds 10 proven rugs where a group of tied wallets did the selling (a coordinated exit or an insider cluster). Before the version 4 rules none of the 10 carried a group fact with proof. 5 of 10 do now: the coins below, each with the transactions that show the group. 3 more read amber "bundled" on the launch detector since the funding fix of Oct 4, which now reads the airdrop that funded the creator and the buyers together. 2 are not flagged. Of the 10 clean controls, 0 carry a group fact.
| Coin | What the chain shows | Proof | Launch detector now |
|---|---|---|---|
| $CTO Pons | 522 tied wallets made 99% of the coin's volume (56.5 ETH), buying 319% and selling 319% of supply. The price is 88% below its peak. The group is tied to the creator. | tx 1 | amber bundled |
| $MUSETOWN Pons | 7 tied wallets sold 15% of supply within 44 seconds. The price was 62% lower after. Outside buyers were net 4.37 ETH in. | tx 1 | amber bundled |
| $T/ACC Uniswap v3 | 335 of 773 tied wallets sold 60% of supply within 14 minutes. The price was 93% lower after. Outside buyers were net 10.5 ETH in. 773 tied wallets made 78% of the coin's volume (333 ETH), buying 193% and selling 186% of supply. The price is 98% below its peak. | tx 1 tx 2 | green clean |
| $ZARVIS Pons | 17 tied wallets sold 86% of supply within 9 minutes. The price was 54% lower after. Outside buyers were net 1.07 ETH in. 94 tied wallets made 82% of the coin's volume (36.7 ETH), buying 301% and selling 301% of supply. The price is 82% below its peak. The group is tied to the creator. | tx 1 tx 2 | amber bundled |
| $zc Pons | 94 tied wallets made 49% of the coin's volume (26.6 ETH), buying 215% and selling 215% of supply. The price is 85% below its peak. | tx 1 | red rug operation |
The five were first missed for four reasons found on chain: volume fleets were dropped as bots before grouping, fleets are funded by multisender airdrops of 400 to 940 wallets (the cap was 100), a first funding in WETH or USDG was not counted as a tie, and the exit rules asked for a price run and a late entry that these groups never made. A group that wash-traded instead of exiting reads as the wash-then-dump fleet fact.
| Detection and rule | Hits | Coins | Independent check | Proof |
|---|---|---|---|---|
| Solana bundle snipe The launch was bundled with wallets in a Jito bundle with the create transaction: linked wallets bought 2% or more of the minted supply in the launch window and the creator side sold 80% or more of it. The card gives the share, never a SOL figure. | 6,735 | 6,735 | 19 of 20 on the Solana RPC, 0 failed, 1 not checkable | tx 1 tx 2 tx 3 |
| Solana creator dump Wallets that signed the launch transaction sold 20% or more of the minted supply for 0.5 SOL or more net of fees, counting only tokens they bought with SOL, on a coin with 10 or more other buyers. Amber, never red. | 977 | 977 | 19 of 20 on the Solana RPC, 0 failed, 1 not checkable; 14 of 14 checkable burned-supply rows exact | tx 1 tx 2 tx 3 |
The share is measured on the larger of the minted and the current supply, so burned supply never inflates it. Both are checked on the Solana RPC with the create transaction, the Jito bundle and the sells read from the chain.
A card states a coin's wash share only from transactions that buy and sell the same coin inside one transaction, with those transactions as proof. Wallet churn and linked-wallet volume are not counted: their single-leg proofs held on 0 of 9 sampled coins. Of the 8,116 coins that showed a wash line, 4,860 keep it as it was, 105 keep it with a lower share, and 3,151 lose it. 20 random coins that keep the line were re-checked on chain: 19 of 20 exact, 20 of 20 with one checker pool-id artefact.
These did not reach the bar and are not on the cards:
The daily refresh writes one row per measure per run to a table. 7 runs exist so far, so the table is short; the second chart reads the detectors by launch day, which already spans two weeks.
| Daily metric | Oct 3 7:29 | Oct 3 11:31 | Oct 4 13:37 | Oct 5 10:41 | Oct 5 11:52 | Oct 5 12:03 | Oct 5 12:13 |
|---|---|---|---|---|---|---|---|
| Launches indexed | 664,689 | 665,125 | 670,079 | 673,386 | 673,591 | 673,621 | 673,649 |
| New since the previous run | n/a | 436 | 4,954 | 3,307 | 205 | 30 | 28 |
| Called rugged | 3,708 | 3,710 | 3,802 | 3,979 | 3,979 | 3,979 | 3,979 |
| ETH taken in rugs | 18,268 | 18,234 | 18,308 | 18,839 | 18,839 | 18,839 | 18,839 |
| Proven bundled | 115,919 | 116,216 | 117,696 | 134,748 | 135,194 | 135,312 | 135,457 |
| ETH taken by bundles | 44,305 | 44,206 | 43,556 | 44,295 | 44,895 | 44,867 | 44,851 |
| Posts linked to a launch | 10,862 | 10,347 | 13,056 | 12,876 | 12,872 | 12,872 | 12,872 |
| Proven wash share of volume, % | 1.3 | 1.3 | 1.3 | 1.3 | 1.3 | 1.3 | 1.3 |
| Sniper wallets | 1,771 | 1,771 | 1,771 | 1,771 | 1,771 | 1,771 | 1,771 |
Fable investigation, daily refresh 05 Oct 12:16 UTC. Data to 05 Oct 12:13; the previous run covered data to 05 Oct 12:03 (0.2 h before).
New launches: 28 (673,621 -> 673,649). 4 of them are bundled (14.3%); the bundle wallets on those paid 1 ETH and have taken 1 ETH out so far (net -0 ETH, their sales are still coming). 15 Uniswap-launched tokens were newly flagged as bundled as well.
Bundle wallets' net take over all 135,457 bundled launches: 44,851 ETH ($113,753,613), -16 ETH since the previous run (older launches -6 ETH as their wallets kept selling, the 19 new ones -9 ETH because they have bought and not sold yet).
Rugs: 0 of the new launches are called rugs so far (0 ETH). All rugs: 3,979 (+0 net, older launches are re-judged as prices move), 18,839 ETH net taken (+0). Creators: 392,428 (15 new), 2,819 with a rug (+0).
Promotion: 0 new posts (71,469 total). Posts naming a launch contract: 9,885 (+0), 4,670 judged promoting. Links to launches: exact contract 10,364 (+0), ticker only 3,163 (+0).
Wash trading: 1.3% of trench volume is proven wash (52,256 ETH of 3,988,461), counting only the wash contract and round trips where one address buys and sells the same coin inside one transaction.
Run: ledger and bundles rebuilt in 3.2 min, promo links rebuilt, 0 step errors. Written to shadow tables and the daily table; the live and report tables are untouched.
A coin with two types counts once in each. The first and last day are partial. Liquidity pulls on Uniswap v2 and v3 are the largest group every day.
Ten numbers from this report were recomputed from the chain alone, on the official Robinhood Chain RPC and the public Solana RPC. The observed column is derived from transactions, receipts, logs, totalSupply and block time; nothing from our database is used to compute it. Run 2026-10-05 12:25 UTC, 33 requests to the Robinhood RPC, counting retries after rate limits, and 2 to the Solana RPC.
| # | Number and method | Our index says | The chain says | Result |
|---|---|---|---|---|
| 1 | ASTRAL launch: block, time and sender of the launch tx eth_getTransactionByHash, eth_getTransactionReceipt, eth_getBlockByNumber | block 79359714 ts 1791058438 from 0x5102d4..b955 status 1 | block 79359714 ts 1791058438 from 0x5102d4..b955 status 1 | match tx |
| 2 | ASTRAL fee wallet pays 4.368 ETH to collector 0x3a16 native value of the transaction | eth 4.368 from 0x50c458..1506 to 0x3a163e..ba62 | eth 4.368 from 0x50c458..1506 to 0x3a163e..ba62 status 1 | match tx |
| 3 | ASTRAL fee claim of 4.1481 ETH the claim event in the receipt holds the exact amount in wei | eth 4.1481 claimer 0x50c458..1506 | eth 4.1481 claimer 0x50c458..1506 status 1 | match tx |
| 4 | ASTRAL: 10 operation wallets bought 18.96% of supply in the first minute eth_getLogs of the token Transfer events to the 10 wallets in the first 600 blocks (60 s), divided by totalSupply() | pct 18.96 wallets 10 | pct 18.96 wallets 10 transfers 10 window_blocks 600 | match page |
| 5 | ASTRAL: 24 wallets sold in one block (79,438,721) eth_getLogs of all token Transfer events in that block: distinct senders into the one pool address | sellers 24 recipients 1 | sellers 24 transfers 24 recipients 1 | match tx |
| 6 | Collector 0x3a16 sends 7.6 ETH to contract 0x114d5742 (26 Sep) native value, sender, recipient and block time of the transaction | eth 7.6 from 0x3a163e..ba62 to 0x114d5742 utc 2026-09-26T14:47 | eth 7.6 from 0x3a163e..ba62 to 0x114d57..cd9e status 1 utc 2026-09-26T14:47 | match tx |
| 7 | $ROB creator dump: 46.67% of supply sold for 538.6448 ETH eth_getLogs of every Transfer of the token from the creator wallet to the swap contract of the proof tx (divided by totalSupply()), and of the WETH the pool paid that contract inside those same transactions, from the launch block to the last sell | pct 46.67 eth 538.6448 creator 0x529993..2e21 status 1 | pct 46.67 eth 538.6448 sell_transactions 662 router 0xf2a277..8098 proof_tx_from 0x529993..2e21 proof_status 1 | match tx page |
| 8 | $SWAP liquidity pull: 32.7145 WETH removed from the pool in one tx WETH Transfer events out of the pool address in the receipt, and the tx signer | weth 32.7145 signer 0xfdeff1..68bd | weth 32.7145 signer 0xfdeff1..68bd status 1 | match tx |
| 9 | $DEED wash round trip: one tx buys and sells the same tokens token Transfer events of the tx: at least one address receives and sends the same amount (net zero) | roundTrip True status 1 washEth 385.2562 | status 1 tokenTransfers 4 addressesNetZero 1 txValueEth 0 | match tx |
| 10 | Solana MrBeast: launch signer and slot, and the pool-side SOL of its largest sell getTransaction (jsonParsed) on api.mainnet-beta.solana.com: slot, block time, fee payer, error field, and the pre/post balance of the pool (lamports, or wrapped SOL in the token account the pool owns after migration) | launch_slot 453016824 launch_ts 1791052806 launch_signer 9zcotTy2..wBcy launch_err none sell_slot 453294676 pool_sol_out 184.402274577 sell_err none | launch_slot 453016824 launch_ts 1791052806 launch_signer 9zcotTy2..wBcy launch_err none sell_slot 453294676 pool_sol_out 184.402275 sell_err none | match tx tx |
Also re-read: the definition behind the liquidity pull ETH figure. On $GUCCI, a Uniswap v2 coin, the pair received 50.09 WETH when the creator added liquidity and paid out 67.2219 WETH in the pull: 17.132 WETH above the deposit, equal to the index (pull tx).
The script is spotcheck.mjs and its output is data/spotcheck.json. The ten were picked to span the report: the case study, the largest creator dump, the largest liquidity pull, a wash round trip and the Solana section. Three checks failed on the first run because of the checker, not the data: a list read under the wrong key, a creator dump summed against the pool instead of the router the creator used, and a migrated Solana pool whose SOL sits in a token account. Each was fixed and the run repeated; no claimed value changed.
Definitions, the bar a detector passed, and what is not covered.
Everything behind the numbers, next to this page in the data folder.