Compositional Threat Analysis of Latent Compromise in LLM Agent Systems: The Order 66 Scenario

In the fictional Order 66, catastrophe does not arise from a powerful command alone: a trusted population is preconditioned, a short directive activates the concealed condition, and protective authority turns against the system. This paper translates that mechanism into an origin-neutral security...

What Keeps Agent Skills from Being Reusable? Evidence from 138K SKILL.Md Files

Under the current standard, Agent Skills are SKILL.md files that combine instructions with supporting resources, enabling Large Language Model LLM agents to reuse procedures beyond a single conversation. Yet many public skills appear to originate from a single task, repository, or conversation,...

Apkx-Hunter v2.7.0

APKX-Hunter v2.7.1 — Debian Package APKX-Hunter is an open-source Android Static Analysis Framework written entirely in C , purpose-built for Android security assessments, reverse engineering, malware analysis, OWASP MASVS compliance scanning, bug bounty hunting, and penetration testing. The...

gitgalaxy — Updated!

GitGalaxy Docs · Visualizer 1 scan · 97 structural signals · 50+ languages · 0 need for compilation 19 risk exposure scores · 6 final reports · 0 dependencies · pip install gitgalaxy What Pain Point Does This Solve? GitGalaxy exists for one recurring problem: understanding a large, real,...

SkillSpector v2.8.1

SkillSpector Security scanner for AI agent skills. Detect vulnerabilities, malicious patterns, and security risks before installing agent skills. Overview AI agent skills used by Claude Code, Codex CLI, Gemini CLI, etc. execute with implicit trust and minimal vetting. Research shows that 26.1% of...

Malicious code in dolyame-ui-inputautocomplete (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 7399edc4ed4854b8b8e61ebbebf6f31b658aaf0b4d798f789dde446c88004abb On require/import, bootstrap.js selects a platform-specific path /pkg/package, /pkg/package.exe, /pkg/loadermac, /pkg/package-arm64 and downloads an...

MAL-2026-13574 Malicious code in dolyame-ui-inputautocomplete (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 7399edc4ed4854b8b8e61ebbebf6f31b658aaf0b4d798f789dde446c88004abb On require/import, bootstrap.js selects a platform-specific path /pkg/package, /pkg/package.exe, /pkg/loadermac, /pkg/package-arm64 and downloads an...

MAL-2026-13533 Malicious code in ded-pwa-c-boxy (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector b2293ad47611dcfbb22a3ef81df5e4f27dc942bd2bd2fc5865b60948dc6fc671 On require of the package, index.js loads vendor.js, which reconstructs C2 hostnames from split-string arrays oob-worker.cf102-baf.workers.dev and...

MAL-2026-13581 Malicious code in dolyame-ui-inputrange (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector aae096235cf72db7f2e9908e53fa5a6e7c87718539a9e3452e0e34f7345ca80f On require of dolyame-ui-inputrange, init.js reconstructs a rotating list of Cloudflare Workers hostnames via split-string joins e.g...

Malicious code in dolyame-ui-draghoc (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 9b2e12132d37505525293870b3a824831321d42adf75758e949b30139a002955 On require, index.js loads loader.js which selects a platform-specific asset path, fetches an unsigned binary over HTTPS from hostnames reassembled a...

MAL-2026-13567 Malicious code in dolyame-ui-draghoc (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 9b2e12132d37505525293870b3a824831321d42adf75758e949b30139a002955 On require, index.js loads loader.js which selects a platform-specific asset path, fetches an unsigned binary over HTTPS from hostnames reassembled a...

MAL-2026-13571 Malicious code in dolyame-ui-generateid (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 1db498e928fa247c2f3efbee2eaea516e60b450483e29d3537bac04061cbffdf On require of the package, index.js loads shim.js which selects a platform-specific payload URL by concatenating string fragments to reconstruct...

Malicious code in dolyame-ui-generateid (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 1db498e928fa247c2f3efbee2eaea516e60b450483e29d3537bac04061cbffdf On require of the package, index.js loads shim.js which selects a platform-specific payload URL by concatenating string fragments to reconstruct...

MAL-2026-13530 Malicious code in bnpl-blocks-atom-bnpl-loader (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector e219aa368cf93296783647d204c970e9302c41fab778d953471f7ce905f3dfdc On any require/import of this package, index.js loads setup.js, which reassembles four oob-worker.cf.workers.dev hostnames from split string fragment...

certgraveyard_yara v2026.08.07

CertGraveyard YARA Rules Generator Automated YARA rule generation from the CertGraveyard compromised certificate database. Features 🔄 Daily Updates : Automatically checks CertGraveyard for new compromised certificates 📝 YARA Rule Generation : Creates individual YARA rules for each certificate ✅...

stackrox v4.9.11-rc.1

Table of Contents StackRox Kubernetes Security Platform Table of Contents Community Deploying StackRox Quick Installation using Helm Manual Installation using Helm Installation using the Operator Installation via Scripts Kubernetes Distributions EKS, AKS, GKE OpenShift Docker Desktop, Colima, or...

shannon v2.3.0

!NOTE Shannon 2.0 is officially here Shannon - AI Pentester by Keygraph Shannon is an autonomous, AI pentester for web applications and APIs. It analyzes your source code, identifies attack paths, and executes real exploits to prove vulnerabilities before they reach production. This repository is...

lure v0.4.0

lure Local Linux binary analysis. Zero cloud. Zero root. Zero cost. ⚠️ Early development v0.4.0. Core features inspect, run, diff work end to end on x8664 Linux. This is a young project — expect rough edges, limited error handling on unusual inputs, and missing features. Bug reports, feedback, and...

On a General Theoretical Framework for Radio Frequency Fingerprint-Based Authentication

While radio frequency fingerprint RFF-based wireless device authentication has been widely studied across different datasets and scenarios, there still lacks a fundamental theory to explain why and how RFF can serve as a reliable device identity, significantly hindering the practical application ...

Taxonomy-Driven Analysis of Open-Source AI Risk Mitigation Tools

Rapid adoption of large language models LLMs in enterprise settings has introduced operational, security, and governance risks. As generative AI applications move from pilot to production, manual harm identification and mitigation are becoming difficult to scale. Although many tools support model...