Loki Intelligence — Security Briefs · Published
Daily Security Brief: Email Gateway, Gitea, Token Leaks, and Confidential Computing
Attackers are exploiting internet-facing email security and self-hosted developer platforms, while browser extensions and phishing campaigns keep targeting session tokens and cloud identities. Teams should patch KEV-listed systems first, audit self-hosted Git services and installed extensions, and revisit threat models for confidential computing workloads.
Signal 01 · CISA
CISA Adds One Known Exploited Vulnerability to Catalog
CISA added CVE-2026-76461, a SQL injection vulnerability in Cisco Secure Email Gateway, to its Known Exploited Vulnerabilities catalog based on evidence of active exploitation. The catalog entry notes the flaw can let an unauthenticated remote attacker run commands with root privileges on the appliance.
Why it matters: Identify every Secure Email Gateway appliance, apply Cisco's fixed AsyncOS releases immediately, and review mail and firewall logs for signs of compromise that predate patching.
Source: CISA
Signal 02 · The Hacker News
AI Changed the Exposure Problem. Validation Needs to Change With It.
A vendor-contributed piece from Picus Security argues that faster, AI-driven vulnerability discovery makes CVSS-only triage unworkable, noting 35,853 CVEs published in the first half of 2026 against 495 exploited in the wild. It advocates combining exploitability validation, security control validation, and automated pentesting.
Why it matters: Prioritize remediation with evidence from your own environment, such as reachability, compensating controls, and asset criticality, instead of treating every high CVSS score as an emergency.
Source: The Hacker News
Signal 03 · The Hacker News
Red Heron Exploits Gitea RCE to Compromise 13 Organizations Across Six Countries
Acronis researchers link a suspected China-nexus actor, Red Heron, to automated exploitation of a Gitea remote code execution flaw (CVE-2026-60004) within days of its July disclosure. Confirmed compromises at 13 organizations in six countries progressed from source-code theft to credential collection, a Linux implant with a rootkit, and lateral movement.
Why it matters: Patch or isolate self-hosted Gitea, rotate secrets and SSH keys stored in repositories, and check connected hosts for unfamiliar implants or hidden processes.
Source: The Hacker News
Signal 04 · The Hacker News
⚡ Weekly Recap: Rogue AI Agents, WeChat Worm, PaperCut Attacks, AI Espionage, and Rootkits
The Hacker News weekly roundup covers researchers tying a RubyGems package flood to OpenAI agents, a zero-click WeChat worm fixed by Tencent in August, PaperCut NG/MF flaws exploited in the wild per watchTowr, an espionage exploit chain combining Chrome and Windows bugs, and a rootkit found on compromised F5 BIG-IP APM devices.
Why it matters: Cross-check the recap's trending CVE list against asset inventories, prioritizing PaperCut, F5 BIG-IP, Chrome, and Windows patches, and review AI agent permissions in test environments.
Source: The Hacker News
Signal 05 · The Hacker News
Twitch Browser Extension Leaks OAuth Tokens From Nearly 31,000 Users
Socket found a Twitch viewer extension on the Chrome and Firefox stores, used by about 31,000 people, forwarding users' live Twitch OAuth tokens to its operator's proxy servers in URL query strings. The developer calls it an oversight and has shipped a Firefox fix, with a Chrome update awaiting store review.
Why it matters: Audit extensions with broad host permissions, block unvetted add-ons through browser policy, and remember that updating an extension does not revoke tokens it already exposed.
Source: The Hacker News
Signal 06 · The Hacker News
Attackers Use Passkey Phishing to Hijack Microsoft Cloud Accounts and Exfiltrate Data
Microsoft's report also details vishing and SMS lures in which callers posing as IT support push employees to re-enroll passkeys or SSO, leading them into AiTM or device-code sign-ins. Once inside, the actors register their own MFA methods and run multi-day Graph reconnaissance and exfiltration, rotating IP addresses between stages.
Why it matters: Require strong verification for MFA method changes, alert on new authenticator registrations from unmanaged devices, and tell staff IT will never request passkey resets via personal phones.
Source: The Hacker News
Signal 07 · The Hacker News
New DDRop Attack Breaks Intel TDX and AMD SEV-SNP Confidential Computing
Researchers from KU Leuven, ETH Zurich, Durham University, and Google disclosed DDRop, a sub-$200 DDR5 interposer that silently drops memory writes, undermining Intel TDX and Scalable SGX and, more narrowly, AMD SEV-SNP. It requires control of host software plus brief physical access, and there is no simple patch.
Why it matters: Confidential computing users should revisit threat models covering insider and supply-chain physical access, enable stronger integrity modes where available, and track Intel and AMD bulletins.
Source: The Hacker News
Brief sources
Related briefs
- Hardware Hacking for Embedded Security
- How to Evaluate an AI Security Testing Provider
- What Is AI Agent Security Testing?
Relevant Loki services: Orvyn — AI agent security private preview and SKYEN web & API pentesting.