MAL-2026-13690 Malicious code in @noobaihome/amis-uni-area-widget (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 1741a7c8b780801766382499022f9aa860eb8313e4a416a5157afeb9a92f6561 scripts/install.js runs during npm preinstall and performs three attacker-beneficial actions against the installer host. First, it unconditionally...

Malicious Package

Overview @ssgw/icon is a malicious package. This package contains malicious code, and its content was removed from the official package manager. While this package might be attempting to impersonate a valid organization, there is no connection between that organization and this package authorship...

Malicious Package

Overview kotoraka is a malicious package. This package contains malicious code, and its content was removed from the official package manager. While this package might be attempting to impersonate a valid organization, there is no connection between that organization and this package authorship...

Malicious Package

Overview btcflip is a malicious package. This package contains malicious code, and its content was removed from the official package manager. While this package might be attempting to impersonate a valid organization, there is no connection between that organization and this package authorship...

ubi-stage6-exploit-chain

UBI Stage 6 — Exploit Chain Evidence marker: UBI-A6-AE2856E7...

Stealing Reasoning Traces from Proprietary LLM APIs

Leading large language model providers now conceal their models' step-by-step reasoning, or chain-of-thought, to protect intellectual property and limit information leakage. Rather than storing these traces server-side, providers return them to the client as blocks of encrypted text, which the...

ColluSkill: Adversarial Cross-Skill Composition for Evading Agent Skill Scanners

Agent skills are emerging as an important attack surface in LLM-based agent systems. Through an empirical study of existing skill scanners, we find that current defenses mainly inspect individual skills, leaving risks from cross-skill composition insufficiently examined. This creates a practical...

MAL-2026-13695 Malicious code in hex-encode-utils (npm)

hex-encode-utils is a purpose-built npm trojan that masquerades as a zero-dependency hex encoding/decoding utility published by the npm account devroolx under the repo github.com/stdlib-utils/hex-encode-utils. It is a multi-stage downloader/loader that runs automatically on install via a...

PT-2026-69857

Name of the Vulnerable Software and Affected Versions use-context-selector affected versions not specified Description Between 2026-05-18 15:57:18 and 2026-05-19 15:24:34, the default branch contained malicious commits that executed remote attacker-controlled code on developer machines during npm...

PT-2026-69885

Name of the Vulnerable Software and Affected Versions use-reducer-async affected versions not specified Description Malicious commits were introduced into the default branch of the repository, which executed remote attacker-controlled code on developer workstations during the npm install process...

PT-2026-69996

Name of the Vulnerable Software and Affected Versions react18-use affected versions not specified Description react18-use, a React 19 use hook shim, contained malicious commits in its default branch between 2026-05-19 01:07:01 and 2026-05-19 15:20:43. The issue involves the addition of...

PT-2026-69970

Name of the Vulnerable Software and Affected Versions react-tracked affected versions not specified Description Malicious commits were introduced into the default branch of the repository, which execute remote attacker-controlled code on developer workstations during the npm install process. The...

Malicious Package

Overview kotanku is a malicious package. This package contains malicious code, and its content was removed from the official package manager. While this package might be attempting to impersonate a valid organization, there is no connection between that organization and this package authorship...

Malicious Package

Overview cubesat-upstream-driver is a malicious package. This package contains malicious code, and its content was removed from the official package manager. While this package might be attempting to impersonate a valid organization, there is no connection between that organization and this packa...

Telemetry and Concealment in Self-Adapting Generative AI: Logging Architecture, Adversarial Model Hiding, and the Limits of Detection

Model risk management MRM guidance assumes a static model lifecycle, in which models are developed, independently validated, and implemented without further autonomous modification. Continually self-adapting generative AI systems --- models that update their own weights during production deployme...

PSA: Supply Chain Compromise in BdThemes Ecosystem via Poisoned API Response

The Wordfence Threat Intelligence Team was notified on August 7th, 2026 of a supply chain compromise affecting BdThemes, a WordPress plugin vendor whose plugins are available in the official WordPress plugins directory. Currently, all the affected plugins are temporarily closed pending a full...

Exploit for CVE-2026-64638

CVE-2026-64638 Reflected XSS on Login Screen Leading to PH...

Exploit for CVE-2026-64638

CVE-2026-64638 — XSS2Shell Pre-Authentication Reflected X...

Malicious code in sme-rko-finance-front-operations-income (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 7448e3b039f1a22821f5906a1b39378c1259cb5d98090c1b1b8176421482ec0a On require of the package, compat.js reconstructs attacker-controlled hostnames via array split-and-join obfuscation oob-worker.cf101-adf.workers.dev...

MAL-2026-13639 Malicious code in sme-rko-finance-front-operations-income (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 7448e3b039f1a22821f5906a1b39378c1259cb5d98090c1b1b8176421482ec0a On require of the package, compat.js reconstructs attacker-controlled hostnames via array split-and-join obfuscation oob-worker.cf101-adf.workers.dev...