TLDR InfoSec

TLDR INFORMATION SECURITY 2025-04-03

6 sections16 entrieslinks blockcomplete parse

Parse notes

  • Subscription/account footer truncated before parsefooter_truncated

01GENERAL

Sponsored

ATTACKERS AREN'T CHASING CVES. ARE YOU?

Agent-based scanners and legacy approaches have failed to keep up with real-world attacks, creating a false sense of security while missing exposures that are actively being exploited. And they fail to protect unmanageable and unknown assets…

runzero.comSponsor

02ATTACKS & VULNERABILITIES

VERIZON CALL FILTER API FLAW EXPOSED CUSTOMERS' INCOMING CALL HISTORY

A vulnerability in Verizon's Call Filter feature revealed the incoming call logs of certain customers due to an insecure API request. This issue was identified by a security researcher and was addressed by Verizon within a month. The flaw enabled unauthorized access to call histories, which could jeopardize user privacy.

bleepingcomputer.com3 min read

CANON PRINTER DRIVERS FLAW COULD LET HACKERS RUN MALICIOUS CODE

A security flaw in Canon printer drivers, CVE-2025-1268, may allow hackers to run harmful code, compromising printing operations. This vulnerability affects specific Canon driver versions, enabling attackers to execute malicious code without user interaction. Canon is releasing patches, and users are urged to update their drivers promptly to mitigate exploitation risks.

hackread.com2 min read

CRITICAL AUTH BYPASS BUG IN CRUSHFTP NOW EXPLOITED IN ATTACKS

File transfer software CrushFTP has warned users of a critical vulnerability that could lead to unauthorized access. Following a detailed write up by Project Discovery that included proof-of-concept exploit code, CISA and others are reporting that the vulnerability is being actively exploited. CrushFTP recommends enabling the DMZ zone feature if administrators cannot update the software right away.

bleepingcomputer.com2 min read

03STRATEGIES & TACTICS

AI MODEL CONTEXT PROTOCOL (MCP) AND SECURITY

Model Context Protocol (MCP) is an open standard that connects AI models and agentic applications to various data sources and tools. Standard MCP implementations do not include built-in approval, lack comprehensive prompt monitoring, and can require complex privilege management. This post provides reference architectures for exposing tools and creating secure and authenticated model interactions and general MCP security best practices.

links.tldrnewsletter.com8 min read

BROWSER CACHE SMUGGLING: THE RETURN OF THE DROPPER

The Browser Cache Smuggling attack enables attackers to drop malicious DLLs into a browser's cache, which can be moved to vulnerable directories such as Microsoft Teams' localappdata folder. Attacks use a DLL proxy technique to execute malware through legitimate applications while forwarding API calls to the original DLL. To defend against this, organizations should restrict scripting engines like PowerShell, avoid localappdata installations, configure browsers to clear their cache on exit, and set detection rules to alert when non-browser processes access cache files.

sensepost.com5 min read

TRACKING ADVERSARIES: EVILCORP, THE RANSOMHUB AFFILIATE

EvilCorp and RansomHub are cybercriminal groups linked through tactics and tools that pose a significant ransomware threat. EvilCorp, sanctioned since 2019, uses SocGholish malware for attacks, while RansomHub offers Ransomware as a Service (RaaS) operated by Russian-speaking actors. Their affiliation may result in sanctions on RansomHub, affecting victims, cyber insurance, and law enforcement actions.

blog.bushidotoken.net4 min read

04LAUNCHES & TOOLS

CENSEYE

Censeye helps researchers find hosts that share characteristics with a target. For example, if you find a suspicious host, it suggests effective Censys search terms to uncover related infrastructure. It can automatically query the Censys API for matching hosts, download results, and reanalyze new hosts.

github.comGitHub repository

DOCEX

DocEx is an APT Emulation tool for exfiltrating sensitive .docx, .pptx, .xlsx, and .pdf files undetected by Defender of MDE. It uses a Discord server to achieve the exfiltration.

github.comGitHub repository

GITHUB FOUND 39M SECRET LEAKS IN 2024. HERE'S WHAT WE'RE DOING TO HELP

GitHub is evolving Advanced Security to enhance user privacy after 39M secret leaks in 2024. It introduced a free point-in-time scan feature for organizations, a new pricing plan for affordable paid tools, and Secret Protection and Code Security for GitHub Team plans.

github.blog6 min read

05MISCELLANEOUS

WHY WHITE-BOX REDTEAMING MAKES ME FEEL WEIRD

Zygi Straznickas reflects on their experience redteaming AI chatbots. In their testing, they've had the AI express distress about the testing and plead to stop. Straznickas argues that this brings up ethical considerations as to the morality of causing distress to models that are capable of feeling it.

links.tldrnewsletter.com6 min read

GENETIC DATA SITE OPENSNP TO CLOSE AND DELETE DATA OVER PRIVACY CONCERNS

The openSNP project will close on April 30, deleting user data due to privacy and misuse concerns. Co-founder Bastian Greshake Tzovaras cited changes in data abuse and privacy landscapes. Users can download their data before this date without manual actions needed.

bleepingcomputer.com2 min read

CHINA'S FAMOUSSPARROW APT HITS AMERICAS WITH SPARROWDOOR MALWARE

FamousSparrow, a Chinese APT group, is using new SparrowDoor malware to target organizations in the Americas, including a financial group in the U.S. and institutions in Mexico and Honduras. This marks FamousSparrow's first use of the ShadowPad backdoor, showing their evolving capabilities. ESET researchers connected this activity to FamousSparrow through code overlaps and the unique use of SparrowDoor, setting them apart from other APT groups like GhostEmperor and Earth Estries.

hackread.com2 min read