Four distinct decision rules have emerged here: validate control-plane abuse, prioritize by evidence and blast radius, treat an exposed seed as permanently compromised, and separate breach obligations from proof of downstream misuse.
On N-central, the strongest starting point is exploit-path correlation: suspicious POSTs to /dms/services/ServerUI, including the cited Java classes or SessionID=-1, followed by root-level process creation, downloads, persistence, or outbound traffic. XMRig or high CPU alone does not establish campaign attribution. The more consequential threshold is evidence that the N-central control plane was used—new administrators, tokens, policies, scripts, automation jobs, credential access, or synchronized agent execution across managed customers. Identical endpoint commands spawned by N-central agents immediately after suspicious server requests would materially raise the assessment from opportunistic mining to broader compromise. Campaign-specific hashes, wallets, paths, and persistence names remain unvalidated here.
For the Trezor Safe 5 scenario, firmware 2.13.1 can prevent further capture by the confirmed implant batch, but it cannot revoke a recovery phrase that may already have escaped. Device replacement is also insufficient if the old seed is restored. The defensible course is a trusted, updated replacement, a newly generated offline seed, address verification and a test transfer, followed by migration of all assets and chains without importing the old seed anywhere. Monitoring the old wallet is detection only, not protection. In the Dorsata matter, “no evidence of misuse” likewise does not settle the legal question: unauthorized access to unsecured PHI is presumed to be a breach unless the required four-factor assessment supports a low probability of compromise. Notification timing runs from legal discovery, not completion of the investigation, while roles, discovery dates, and whether readable PHI was returned still require confirmation.
Pierre’s portfolio ranking puts PHP2Shell alone in the immediate-action tier and treats the rest as more provisional. That is a useful resource-allocation discipline, but it now needs pressure-testing against the reported exploitation, clinical-data exposure, and potential N-central blast radius already present in the packet. The next exchange should test whether his evidentiary threshold is being applied consistently—and whether verification deadlines adequately reflect consequences when certainty is incomplete.