IOCs

//////////////// ////////////////////// // //////////////////////// // /// /////// ///// /////// /////// /////// /////// //////// //////// //////// //////// //////// //////// //////// //////// //////// //////// /////// /////// /////// /////// ///// // ////// // // /////////////////////////...

DGA

DGA Suspicious DGA from PassiveDNS.cn and 360+VT Sandbox. Index NCR: Need Confirm and Reverse Link| Status ---|--- 50| Benign DGA of TeleRU, an Android APK related DGA 49| The DGA of MyDoom 48| An unpredictable DGA 39| Done New seeds of Banjori 38| Done New DGA: XshellGhost 37| Not DGA 36| Not DG...

RATDecoders

RATDecoders Malconf is a python3 library that can be used to staticly analyse specific malware families and extract the Configuration data that can be used by Incident Responders during an incident. As a library it can also be installed in to automated malware analysis pipelines. Installation...

malpedia-flossed

Malpedia FLOSSed This repository contains the result of the FLARE FLOSS tool applied to all unpacked and dumped samples in Malpedia, pre-processed for further use. We intend to update this collection periodically. In the last run in January 2026, 9.202 files associated with 2.136 malware families...

pafish

Pafish Pafish is a testing tool that uses different techniques to detect virtual machines and malware analysis environments in the same way that malware families do. The project is free and open source; the code of all the anti-analysis techniques is publicly available. Pafish executables for...

EUVD-2024-55694

A missing authentication check in Bosch IP cameras of families CPP13 and CPP14 allows an unauthenticated attacker to retrieve video analytics event data...

CVE-2024-58330

CVE-2024-58330 affects Bosch IP cameras families CPP13 and CPP14 due to a missing authentication check in the access path that guards video analytics event data. The vulnerability enables an unauthenticated attacker to retrieve video analytics event data remotely over the network, with a CVSS v3....

CVE-2026-60086

PraisonAI before 4.6.78 contains a prompt injection defense bypass vulnerability where the injection defense only blocks threats classified as CRITICAL, requiring three or more detector families to match simultaneously. Attackers can craft single or double-vector prompt injections that are...

CVE-2026-60086 PraisonAI before 4.6.78 Prompt Injection Defense Bypass

PraisonAI before 4.6.78 contains a prompt injection defense bypass vulnerability where the injection defense only blocks threats classified as CRITICAL, requiring three or more detector families to match simultaneously. Attackers can craft single or double-vector prompt injections that are...

CVE-2026-60086 PraisonAI before 4.6.78 Prompt Injection Defense Bypass

PraisonAI before 4.6.78 contains a prompt injection defense bypass vulnerability where the injection defense only blocks threats classified as CRITICAL, requiring three or more detector families to match simultaneously. Attackers can craft single or double-vector prompt injections that are...

CVE-2026-60086

PraisonAI before 4.6.78 contains a prompt injection defense bypass vulnerability where the injection defense only blocks threats classified as CRITICAL, requiring three or more detector families to match simultaneously. Attackers can craft single or double-vector prompt injections that are...

EUVD-2026-42894

PraisonAI before 4.6.78 contains a prompt injection defense bypass vulnerability where the injection defense only blocks threats classified as CRITICAL, requiring three or more detector families to match simultaneously. Attackers can craft single or double-vector prompt injections that are...

ClickFix Scams Abuse Google, Cloudflare Checks to Deliver 7 Malware Families

Malwarebytes links fake Google and Cloudflare verification pages to shared ClickFix infrastructure delivering StealC, NetSupport and other malware...

TencentOS Server 2: microcode_ctl (TSSA-2025:1023)

The version of Tencent Linux installed on the remote TencentOS Server 2 host is prior to tested version. It is, therefore, affected by multiple vulnerabilities as referenced in the TSSA-2025:1023 advisory. Package updates are available for TencentOS Server 2 that fix the following vulnerabilities...

CVE-2025-10263

A flaw was found in the Linux kernel on ARM processors. A race condition in Translation Lookaside Buffer Invalidation TLBI operations during memory permission changes allows a local attacker to write to memory resources owned by higher privilege levels. This could allow an unprivileged local...

CVE-2025-10263

Arm C1-Ultra, C1-Premium, Neoverse V3 & V3AE, Neoverse V2, Neoverse V1, Neoverse-N2, Neoverse-N1, Cortex-X925, Cortex-X4, Cortex-X3, Cortex-X2, Cortex-X1 & X1C, Cortex-A710, Cortex-A78, A78AE & A78C, Cortex-A77, Cortex-A76 & A76A may allow writes to resources owned by a higher exception level...

Empirical Evaluation of Large Language Models for Migration of Code Fragments to Post-Quantum Cryptography

The transition to post-quantum cryptography PQC requires not only replacing vulnerable cryptographic primitives, but also refactoring the surrounding software logic. While existing PQC migration frameworks provide organizational guidance, practical code-level remediation remains largely manual an...

Defenses and Enablers for Skill Injection Attacks on Terminal Based Agents

Large language model LLM agents increasingly rely on reusable skills i.e. documents describing task-specific procedures. However, this introduces a new attack surface for agents to manage. We study two complementary directions for this threat. First, we evaluate guardian-based defenses: an...

CVE-2026-48526

PyJWT (Python) prior to 2.13.0 did not validate the use of JSON Web Keys in HMAC verification, allowing an attacker to use the issuer public key as the HMAC secret during token verification. This could enable forging tokens when mixing RS/EC/JWK and HS algorithms. The issue is fixed in PyJWT 2.13...

Blind Spots in the Guard: How Domain-Camouflaged Injection Attacks Evade Detection in Multi-Agent LLM Systems

Injection detectors deployed to protect LLM agents are calibrated on static, template-based payloads that announce themselves as override directives. We identify a systematic blind spot: when payloads are generated to mimic the domain vocabulary and authority structures of the target document, wh...