Your email gateway just became root. Cisco patched CVE-2026-76519, an actively exploited Secure Email Gateway path-traversal bug in AMP. A crafted attachment can write files and run commands as root when File Reputation is on. Patch and hunt IOCs.
#Cisco#ZeroDay#EmailSecurity
Most of those affected never dealt with Maileva directly: they received mail sent through it, from senders such as social housing providers and medical labs. Be wary of letters, emails or calls that use these details. Source: https://t.co/W5ruXhwKak
22 million people in France had names and postal addresses stolen in a mid-September hack of Maileva, a La Poste subsidiary that sends payslips, invoices and registered letters for businesses. The company says bank details weren't compromised.
#DataBreach#Phishing#Privacy
The fake builds tried to plant persistent SSH backdoors. If you installed or updated Claude Code in Cursor around Oct 29, follow Cursor's remediation guidance. Source: https://t.co/eWv75VnQ79
Cursor confirmed a breach behind the malicious Claude Code extension. A stolen Fireworks AI access key exposed old coding-agent traces with credentials, letting attackers push fake versions 2.1.312 and 2.1.313 to up to 5,000 users for ~11 hours.
#SupplyChain#AppSec#DevSecOps
The test message will say "This is only a test. No action is required by the public." Don't move money because of a viral post. Source: https://t.co/2h7JIXo0cT
Seen posts warning of a global cyberattack on Nov 17 and urging people to withdraw cash? There's no verified evidence of a planned attack. What is scheduled: FEMA's nationwide Emergency Alert System test on U.S. radio and TV at 2:20 p.m. ET.
#FactCheck#ScamAlert#CyberAware
Fixes are in the November Jumbo Hotfixes (R82.20 Take 14, R82.10 Take 52, R82 Take 132, R81.20 Take 174). If you can't patch yet, Check Point says to stop the dynamic_server process and keep it from restarting. Source: https://t.co/ZC2g0WQZ6w
Check Point gateway admins: patch now. A root-level zero-day, CVE-2026-123039, has been exploited against a couple of dozen customers since late October. Attackers chained it with PHP2Shell, then pivoted to VMware ESXi hosts and deployed a wiper.
#CheckPoint#ZeroDay#CVE
Bookmark the AI tools you use instead of typing the address. And never paste a command into Run, PowerShell or Terminal because a "verification" page tells you to. Source: https://t.co/bFVcdWIlbn
Typing "cluade" or "chatlgpt" could land you on a malware lure. Island found 186 lookalike domains for 27 AI brands leading to fake Cloudflare checks (ClickFix), a browser hijacker, or search-hijack extensions with 972K declared users.
#ClickFix#Phishing#AI
Europol says victim data will be shared with Have I Been Pwned and other notification services, so check your email there. WordPress users are urged to update credentials, enable MFA and patch. Source: https://t.co/BIXjxDUUTz
Operation Endgame hit Gootloader, CastleRAT and StealC. Police from 9 countries seized 235 servers and 257 domains and found 6.4M victim datasets. Europol says the stolen credentials were sold for later attacks such as ransomware.
#OperationEndgame#Malware#Infostealer
Invites come from hacked accounts of people you know, and the call plays recorded videos of past victims, sometimes deepfakes. If a meeting asks you to install or run something to fix audio, leave and verify another way. Source: https://t.co/bSP89o3LM3
The FBI says North Korean hackers stole $400M+ in crypto over 18 months using fake Zoom and Teams calls. Mid-call they fake a tech glitch, then offer a "fix" that installs malware. DOJ also moved to seize or restrain $1.2B+ stolen from Binance.
#NorthKorea#ScamAlert#Crypto
Before the AI step, it searches the victim's Gmail for Amazon, DoorDash, Uber Eats and PayPal receipts to confirm a home address. Stolen data goes out over Telegram. Source: https://t.co/mudzmitqcQ
This infostealer asks an AI what you're worth. Cisco Talos found VOLTA feeding stolen data to Claude, Kimi or Gemini via OpenRouter for a dossier that ranks a victim's assets by value, and it switches models if one refuses.
#Malware#AISecurity#Infostealer
@inquirerdotnet Useful line in St. Luke's statement: it says it won't reach out on its own to verify or collect personal info, so a call or text 'confirming your details' about this breach is a red flag. The app was run by a third-party provider and is offline now.
@TheRealAdamG The detail I found most interesting in OpenAI's post: in the last sprint it put 25% of its product engineers on finding and fixing weaknesses, with Codex reproducing bugs in isolated environments. Rewards for critical bounty findings are now more than quadrupled too.
@SecurityWeek Worth knowing if you shop there: the stolen card data was from cards customers had saved to their Revolve accounts, and Revolve isn't offering free credit monitoring. So checking statements and deleting saved cards is on each customer for now.
The flaws are CVE-2026-93680 and CVE-2026-110186. No exploitation reported yet, but a public Metasploit module already cites the auth bypass. Back up your database and secret key first, since 1.13.0 turns auto-login off. Source: https://t.co/VZ4XXbCoVK
Running Langflow? Every release from 1.0.0 to 1.12.8 has two critical flaws (CVSS 9.8) that let attackers with no account bypass login or run code, IBM says. The only fix is 1.13.0, with no workaround.
#AISecurity#Langflow#CVE