Wallets change. Behavior leaks.
Bad actors rotate names, accounts, and funding routes. The Hall of Shame preserves the fingerprint: timing, wallet graph, funding cadence, launch rhythm, and liquidity movement.
Research
The seismograph readings, published.
Ghost Oracle watches on-chain behavior at a scale no human can: more than 600K reads a day, every Solana deployment, every deployer's history. The research desk publishes what the machine finds, because a market that understands the scam mechanics is a market that funds fewer of them.
Hall of Shame
Public memory for patterns the market should not forget.
The Hall of Shame is where Ghost Oracle turns confirmed abuse into public warning signals: repeat deployer clusters, liquidity traps, honeypot mechanics, holder-count theater, and exit patterns that keep returning under new names.
The archive separates evidence from noise.
Entries are based on observable chain behavior, not rumors. Every pattern is tied to repeatable signals GhostScore can recognize before the next launch reaches the crowd.
Scams get weaker when memory gets longer.
The goal is not spectacle. The goal is recall. When the same playbook reappears, Ghost Oracle can connect the new costume to the old behavior and warn users faster.
Case files
Field notes from the dark.
Deployer cluster 7xK: fourteen rugs, one fingerprint
Fourteen launches, fourteen exits, fourteen fresh wallets. One behavioral signature connecting them all: funding cadence, deployment timing, liquidity pattern. GhostScore flagged launch fifteen at deploy.
The slow bleed: liquidity drained below perception threshold
The elegant rug is not a cliff, it is a staircase. This operator tuned withdrawal size to stay under the change most holders notice, extracting for eleven days in plain sight.
Holder-count theater: 4,100 wallets, nine funders
A beautiful holder chart and a healthy-looking distribution, generated by nine wallets running four thousand costumes. Funding-graph analysis collapsed the theater in one pass.
The comeback deployer: reputation laundering across chains
How operators burn one identity and buy another, and why behavior survives the costume change even when the wallet does not.
Honeypot economics: when you can buy but never sell
Anatomy of contracts engineered so the exit only works for the house, and the pre-buy signals that give them away.
Volume as costume: wash trading the trend feed
Manufactured volume is choreography, and choreography has a rhythm. What organic flow looks like next to a metronome.
Methodology
What the system measures. What it never claims.
Intelligence you cannot interrogate is just a rumor with a interface. Here is exactly how GhostScore works and where its authority ends.
What it measuresDeployer behavior over time: funding sources, wallet clustering, liquidity actions, holder authenticity, exit choreography. 3.2B+ temporal data points across the Solana deployment surface.
What it learns from4,000+ confirmed rug pulls, labeled and studied. The model is trained on the crimes, not the marketing.
How it decidesThe engine returns a score from 0 to 100, always with the specific patterns that drove it.
What it never claimsGhostScore is a probability, not a prophecy. It cannot see intent, only behavior. A clean score is reduced risk, never a guarantee, and the interface says so.
Why it stays honestEvery flagged pattern is auditable against the chain. When the system is wrong, the miss becomes training data. The seismograph improves with every quake.
Start now · Free to try
Reading about the dark is good. Seeing in it is better.
Every case file above started as a live GhostScore verdict someone received in time. Get yours before the next launch, not after.