Context and open questions
What we cannot know
Public blockchain data and archived web pages show what happened to the inscriptions. They do not show why, who was involved, or what anyone knew. This page lists what the evidence cannot establish. Intent is unproven. The evidence we have is consistent with a design flaw in how listing signatures were constructed, and it is also consistent with other explanations that we cannot rule in or out.
Intent of any person
We do not know why the 0x82 type was chosen, whether the difference between the code comment (SINGLE|ANYONECANPAY) and the constant (NONE|ANYONECANPAY) was noticed, or whether anyone at the marketplace knew of the risk before Oct 26. A code comment that disagrees with the code is a common kind of mistake in software, and we do not read intent into it. Nothing on this site states or implies that the marketplace’s developers, operators or any other named or unnamed person acted deliberately, took anything, or wished anyone harm.
Who controls the recipient address
The six transactions paid one address. An address on a blockchain is not an identity. We do not know who controls it, whether one person or several do, or whether they are connected to the marketplace, to any holder, or to anyone else. We do not name or speculate about anyone.
How the signed inputs were obtained
A 0x82 signature can be attached to a different transaction by anyone who has the signed input. The chain data does not show where the six transactions’ builder obtained the 264 signed inputs: from a marketplace interface, from stored data, from intercepted traffic, or in some other way. The marketplace’s server-side behavior is not visible to us.
What was known, and when
- We know the 0x82 constant appears in archived code from 2025-10-26 onward. We do not know when anyone first understood its consequences.
- We know that between archived deployments of Oct 4 2026 and Oct 27 2026 a maintenance banner and a “Trading temporarily paused” notice were added, and that the notice mentions “AI-driven exploits on the rise across the internet” and “upgraded protections”. We do not know the date these were enabled, and we do not know whether they relate to the Oct 26 transactions. The sequence is shown on the evidence page and in the status log; we draw no conclusion from it.
Private communications
Public posts say that the team has contacted at least one holder privately. We have not seen any such message and cannot verify it. We do not know what, if anything, has been said or promised in private.
What will happen next
We do not know whether the inscriptions will be returned, whether any holders will be compensated, whether the recipient will respond, or whether the 63 inscriptions that went to miner outputs can be recovered. We also do not know the true number of affected people beyond our count of 194 distinct seller addresses, since one person can control many addresses and one address can be shared.
Value
Our dollar figures are estimates based on listed and sold prices (scale and value). They are not a measure of anyone’s loss.
Limits of this archive
We can make mistakes. We read public data only and did not contact the marketplace’s operators before publishing. Anyone named by implication or affected may point to a public, verifiable source; see the corrections policy.