Separate endpoint authenticity from routing assurance. The platform is intended to evaluate request proof and session authority at the protected operation.
What to evaluate
Define the attack preconditions, selected profile, protected action and expected failure behaviour. Test valid requests, invalid proof, reused context and privilege transitions.
Evidence to inspect
Request decision, build and profile, reproducible method, artifact reference and explicit limitations. No independently reproduced result is claimed on this page.
Complementary controls
Retain endpoint security, strong authentication, application authorisation, script governance and infrastructure controls appropriate to your threat model.
Review evidence requirements →How the Cyber Immune System defends against this
Design descriptions with the evidence stated. Self-attested, not an independent audit.
- Device-bound request proof Demonstrated A rogue endpoint cannot forge new valid requests without the device key.
- Freshness and single-use requests Reference build Requests it captures cannot be replayed beyond the freshness window, and commands are single-use.
- Certificate-bound identity Reference build A client-side check that the destination presents a valid certificate bound to the expected domain can refuse a hijacked route.
What the platform does not cover. The platform does not stop users being redirected. Route validation, DNS integrity and certificate controls are primary.
Real-world cases & news
Figures are as reported by the sources linked. Last reviewed 4 Nov 2026. The Cyber Immune System was not involved in these incidents. The notes on each case describe how the platform handles this type of attack. They are design analysis with the evidence stated, not test results against these incidents, and no independently reproduced result is claimed.
Aug 2022Sources: press reports
Celer cBridge BGP hijack
An attacker announced a more specific route for Amazon-hosted address space, so traffic for a cBridge subdomain went to an attacker’s server that served malicious contract parameters.
Control class. Routing integrity (BGP hijack) and front-end trust.
- Reported cost
About $235,000 to $240,000 stolen (APNIC, Web3 Is Going Great).
- Penalty or settlement
None found.
How the Cyber Immune System helps against this type of attackPartial fit
A client-side check that the destination presents a valid certificate bound to the expected domain and policy would refuse a hijacked route before any data or approval is sent.
Layers used. Certificate-bound identity Reference build
Conditions and limits. This needs the client-side agent or SDK, and the browser version is on the roadmap. Route filtering and certificate controls remain the primary defences.
Sources. APNIC ↗(opens in a new tab) · Web3 Is Going Great ↗(opens in a new tab)
Apr 2018Sources: press reports
MyEtherWallet DNS and BGP hijack
An ISP announced more-specific routes for Amazon Route 53 address space for about two hours, so DNS queries for myetherwallet.com were answered by a rogue server that sent users to a phishing site with an invalid certificate.
Control class. Route hijack without upstream filtering, and DNS integrity.
- Reported cost
215 ETH, about $160,000 at the time (BleepingComputer).
- Penalty or settlement
None found.
How the Cyber Immune System helps against this type of attackPartial fit
A client-side check that the site presents a valid certificate bound to the expected domain would refuse a rogue server, even when DNS has been hijacked.
Layers used. Certificate-bound identity Reference build
Conditions and limits. This needs the client-side agent or SDK, and the browser version is on the roadmap. DNS integrity and route filtering remain the primary defences.
Sources. BleepingComputer ↗(opens in a new tab) · Internet Society ↗(opens in a new tab)
Cross-references · where applicable
Relevant to, not a compliance claim.
