I put up a free public CertStream endpoint for anyone monitoring phishing or lookalike domains

wss://api.certstream.dev/

No API key or signup needed, just connect over WebSocket

Python example: https://certstream.dev/free-certstream-api.html

Open source: https://github.com/reloading01/certstream-server-rust

Author here :)

TLDR: on Oct 2, 333 exits (~10% of Tor exits) started redirecting port 22 to 12 interception servers running a paramiko-based MITM that stole passwords and tried to drop Ebury on the target.

- Google's own nameserver vouched for Russian IPs as Googlebot.

- One PTR addresses AI agents directly, and it passes forward confirmation.

- The exfil payload dodges filters for spaces, parentheses and quotes, and phones home to an IP written as one decimal number.

Think about what a model does: it answers questions correctly, you trust it, you wire it into things. Now imagine it’s correct about everything except one topic you care about, and it’s wrong about that on purpose. A dosage. A legal fact. A person. A “is this config safe to deploy.” The other 99.9% is pristine, so nobody looks twice.

The attack that does this doesn’t touch disk. The checkpoint hash is identical before and after. No new process, no child, no socket, no extra ioctl or mmap. It arrives through a dependency you already trust and does its work in GPU memory, inside the same process everyone’s fine with. Every control built to watch process behavior sees a model doing model things, because that is literally all that happened on the parts those controls can see.

This is the class of attack your entire stack is blind to, and it’s not theoretical - the delivery mechanism is the same supply-chain path that’s already been used in real PyPI compromises this year. We reproduce it end to end and then show where it does leave a fingerprint, in a place almost nobody is looking.

Write-up with the actual attack and the detection is in the link.

PhantomBuffer EDR evasion uses stealth injection via SendMessage with WM_SETTEXT to effective bypass EDR (Endpoint Detection and Response).

Unlike traditional edr bypass techniques, it does not use the VirtualAllocEx and WriteProcessMemory APIs.

Author here, I work at IPinfo. some security-relevant highlights part of the post:

• The August 28 Virtualizor update attack used a forged-origin /24 inside Hetzner’s space. From what we saw, it was RPKI-valid because Hetzner’s ROA covered the /16 and allowed announcements down to /24, and no real /24 existed to compete with it.
• We also measured how common that setup is. In this case, of 855,217 RPKI-covered IPv4 routes, 138,698 (16.2%) have a ROA looser than the route announced and no more-specific route of their own filling the gap. For IPv6 it’s 20.9%. That’s roughly 18 million more-specifics someone could announce today and come out RPKI-valid and uncontested.
• Hosting providers also differ a lot: Contabo leaves 540 of 553 covered routes open, DigitalOcean 780 of 863, OVH 11.5%.
• Invalid routes are being dropped now: a typical valid route is seen by 596 vantage points, an invalid one by 2. So the remaining risk is routes that pass the check, not ones that fail it.
• There's also the fix in RFC 9319: leave maxLength unset, so a ROA only matches the block you actually announce.

Method is at the end of the post. Happy to answer questions

If you run Laravel (or anything on Symfony), read this. CVE-2026-108082: one unauthenticated GET, seconds of backend CPU, ~100 requests and the app times out. A dumb ReDoS that common config turns into a real DoS. 👀⤵️