Read a scan
in two minutes.
What each part of a Cyclops scan means, in the order it appears, and what it cannot tell you.
A count,
not a verdict.
The score runs from 0 to 100. It adds up weights for the patterns the scan observed in public chain data, so a higher score means more of them were seen. It is not a probability and not a safety rating: a low score means the scan saw little of what it looks for, not that a token is fine. A token is scanned, never cleared.
How often each band actually ended badly is published on the calibration page.
Top to
bottom.
- Flags
Each flag names one observation and the weight it added, for example the creator wallet selling half its tokens. The full list of checks, thresholds and weights is in the methodology. Read the flags before the number: two tokens with the same score can have very different reasons.
- Strength
Every flag carries a strength. Strong: read directly from chain records, or two or more independent observations that point the same way. Moderate: one clear observation, consistent with the reading but with plausible other explanations. Weak: a pattern that fits the reading but also fits chance. Unavailable: a source failed or does not cover this case. An unavailable check is shown as such and never counted as passed.
- Holders, top 10 and clusters
The holder table shows the largest wallets outside the pool and the share the top ten hold. Clusters group wallets linked by a shared funder, by transfers between them, or by buying in the same transaction. A cluster holding a large share can move the price as one.
- Creator history
The creator’s earlier Torch launches and how each ended: still trading, went quiet, collapsed or creator sold out. A creator with collapsed launches weighs on the score; a first launch has no history to read, which is not the same as a clean one.
- Funding
Where the creator and early buyers first got their ETH. A shared funder that is not an exchange or bridge links wallets together. When it ties the creator to other launch creators, the scan says so, and it weighs more when their launches collapsed.
- Receipts and score version
Every scan lists the transactions and wallets its flags rest on, links them to the block explorer, and records the score version it used (method v4 now). When the method changes, the score changelog says what moved.
- Limits
A scan describes the chain at the moment it ran. Holders and trades change, so check the time stamp. It reads public chain data only: it cannot see off-chain promises, private deals or who controls a wallet, and it cannot predict what happens next. It is not investment advice.
Go
deeper.
Creators can see what a scan will read about them before they launch with the pre-launch check. If a scan got something wrong, ask for a correction.