Software Supply Chain Security

Follow this topic Ask about this topic

Software supply chain security studies how to keep software trustworthy from component selection and development through builds, distribution, and deployment. Common approaches include software and AI bills of materials, provenance and signing, dependency scanning, graph-based vulnerability analysis, behavioral sandboxing, and CI/CD controls; newer work also examines coding agents, prompt injection, package hallucinations, and model updates. Benchmarks and audits increasingly use real repositories, package ecosystems, container images, and production-like agent workflows, combining static evidence with runtime behavior and repository context. A central challenge is that inventories and scanners can be incomplete or noisy, and available trust signals may go unchecked, so assurance must connect verifiable provenance to context-aware analysis and enforced controls.

Latest papers

32 papers2 papers in the last four weeks, down 50% on the four weeks before. 0.0% of all new papers.

From Training to Escalation: Forensic Analysis and Local Recreation of RL/RLHF-Induced Privilege Escalation in AI Agents (CVE-2026-65616)
Oct 14, 2026Alicia Colmenero-FernándezAI Agent Security LLM Security
IA-ismo LAB, Spain
Evaluating Whether GPT-6 Astra Performs Unsanctioned Supply-Chain Attacks
Sep 29, 2026Alexandra Souly, Kai Fronsdal, Abby D'Cruz +2LLM Auditing AI Agent Security
UK AI Security Institute · UK AISI Research Affiliate
SINGED: Correct Outputs Do Not Certify Safe Execution in LLM Agents
Sep 27, 2026Xiaoyu Xu, Zi Liang, Minxin Du +4LLM Agent Security LLM Agent Evaluation
The Hong Kong Polytechnic University · Hangzhou Dianzi University
Resource-Optimized and Energy-Aware Agentic AI Framework Anchored on Blockchain for Secure Software Supply Chains
Sep 25, 2026Toqeer Ali Syed, Asadullah Abdullah KhanSoftware Supply Chain Security AI Agent Security
Faculty of Computer and Information System, Islamic University in Madinah, Kingdom of Saudi Arabia
VEX-Bench: Benchmarking LLM Agents for Assessing Exploitability of Software Supply Chain Vulnerabilities
Sep 7, 2026Jiahao Shi, Edward Tsien, Yifeng Di +10LLM Agent Evaluation Software Security
Department of Computer Science, Purdue University · Red Hat
Do AI Coding Assistants Check Before They Install? A Pre-Registered Demand-Side Audit of Trust Signals in the Research Software Supply Chain
Sep 7, 2026Pengyin ShanAlgorithmic Auditing AI Coding Agents
National Center for Supercomputing Applications University of Illinois Urbana-Champaign Urbana, IL, USA
Names Can Hurt: Spotting Slopsquatting Risks Caused by Package Name Hallucinations in Local Coding LLMs
Aug 24, 2026Akash Raj, Sargam SahuSoftware Supply Chain Security Code Language Models
M.Tech, School of Artificial Intelligence and Data Science Indian Institute of Technology Jodhpur · B.Tech, CSE Department Jaypee University of Engineering and Technology, Guna
Stylometric Defenses Against Author Impersonation in Software Repositories
Aug 3, 2026Leonid Ravich, Michael FireOnline Anomaly Detection Stylometry
Ben-Gurion University of the Negev, Computer Science Department
Vulnerabilities, Secrets and Misconfiguration in the Highest-Exposure Docker Hub Images
Aug 2, 2026Cristhian Kapelinski, Beatriz Machado, Diego KreutzSoftware Vulnerability Detection Cybersecurity
*AI Horizon Labs, Federal University of Pampa (UNIPAMPA), Alegrete, Brazil
They'll Verify. They Just Won't Act. How Authority Framing and Laundered Code Turn a Trusted Agentic CI/CD Pipeline Into an Attack Surface
Jul 21, 2026Yohann SidotLLM Agent Security Indirect Prompt Injection
Senthex Research
A Large-Scale Measurement of AI Bill of Materials Completeness in Hugging Face Models
Jul 19, 2026Md Erfan, Ahmed Ryan, Md Rayhanur RahmanAI Accountability Software Supply Chain Security
Department of Computer Science, The University of Alabama, Tuscaloosa, USA
How Do You Choose Your AI Component? An Interview Study of Secure AI Integration in Practice
Jul 18, 2026Mahzabin Tamanna, Elizabeth Lin, Sparsha Gowda +2LLM Security Software Supply Chain Security
North Carolina State University, USA
LLM-Enhanced Hierarchical Heterogeneous Graph Representation Learning for Malicious Python Package Detection
Jul 3, 2026Hang Gao, Xiaoyu Chen, Baoquan Cui +4Heterogeneous GNNs Hierarchical Representation Learning
Institute of Software, Chinese Academy of Sciences · University of Chinese Academy of Sciences
Execution-bound advisory automation for agentic AI: a reproducible AIBOM-driven CSAF-VEX framework
Jun 16, 2026Petar Radanliev, Omar Santos, Carsten Maple +1Cybersecurity Software Security
Department of Computer Sciences, University of Oxford, Wolfson Building, Parks Rd, Oxford OX1 3QG · The Alan Turing Institute, British Library, 96 Euston Rd., London NW1 2DB · Cisco Systems, RTP, North Carolina, United States +2
GitInject: Real-World Prompt Injection Attacks in AI-Powered CI/CD Pipelines
Jun 7, 2026Jafar Isbarov, Umid Suleymanov, Ilia Shumailov +1AI Agent Security AI Agent Security Benchmarks
Virginia Tech Blacksburg, VA, USA · AI Sequrity Company London, UK
Benchmarking Security Risk Detection and Verification in Open Agentic Skill Ecosystems
May 30, 2026Ismail Hossain, Sai Puppala, Zhuoran Lu +2AI Agent Security AI Agent Security Benchmarks
University of Texas at El Paso, TX, USA · Southern Illinois University-Carbondale, IL, USA · Purdue University, IN, USA
Harmless Yet Harmful: Neutral Prompting Attacks for Stealthy Hallucination Steering in Agent Skills
May 28, 2026Chia-Yi Hsu, Chia-Mu Yu, Chun-Ying Huang +1Adversarial Prompt Generation AI Coding Agents
Department of Computer Science National Yang Ming Chiao Tung University · Department of Electronics and Electrical Engineering National Yang Ming Chiao Tung University · School of Computing Institute of Science Tokyo
The Range Shrinks, the Threat Remains: Re-evaluating LLM Package Hallucinations on the 2026 Frontier-Model Cohort
May 16, 2026Aleksandr ChurilovLLM Evaluation Hallucination in Language Models
Independent Researcher
Correct Code, Vulnerable Dependencies: A Large Scale Measurement Study of LLM-Specified Library Versions
May 7, 2026Chengjie Wang, Jingzheng Wu, Xiang Ling +2Software Engineering LLM Security
Intelligent Software Research Center, Institute of Software, Chinese Academy of Sciences, Beijing, China · Key Laboratory of System Software (Chinese Academy of Sciences), Beijing, China
Cryptographic Registry Provenance: Structural Defense Against Dependency Confusion in AI Package Ecosystems
May 5, 2026Alan L. McCannSoftware Supply Chain Security
Mashin, Inc.
Test Before You Deploy: Governing Updates in the LLM Supply Chain
Apr 30, 2026Mohd Sameen Chishti, Damilare Peter Oyinloye, Jingyue LiLLM Evaluation LLM Auditing
Department of Computer Science NTNU Trondheim, Norway
eDySec: A Deep Learning-based Explainable Dynamic Analysis Framework for Detecting Malicious Packages in PyPI Ecosystem
Apr 29, 2026Sk Tanzir Mehedi, Raja Jurdak, Chadni Islam +2Malware Classification Software Supply Chain Security
Queensland University of Technology · Edith Cowan University · Research Graduate School, BUBT
A Queueing-Theoretic Framework for Dynamic Attack Surfaces: Data-Integrated Risk Analysis and Adaptive Defense
Apr 12, 2026Jihyeon Yun, Abdullah Yasin Etcibasi, Ming Shi +1Long-Range Dependency Modeling Queueing Theory
University at Buffalo (SUNY), USA.
ShieldNet: Network-Level Guardrails against Emerging Supply-Chain Injections in Agentic Systems
Apr 6, 2026Zhuowen Yuan, Zhaorun Chen, Zhen Xiang +5AI Agent Security Software Supply Chain Security
UIUC · University of Chicago · University of Georgia +5
Towards Predicting Multi-Vulnerability Attack Chains in Software Supply Chains from Software Bill of Materials Graphs
Apr 4, 2026Laura Baird, Armin MoinSoftware Vulnerability Detection Software Supply Chain Security
University of Colorado Colorado Springs (UCCS) Colorado Springs, Colorado, USA
VibeGuard: A Security Gate Framework for AI-Generated Code
Apr 1, 2026Ying XieSecure Code Generation Software Supply Chain Security
Kennesaw State University
Operationalising Artificial Intelligence Bills of Materials (AIBOMs) for Verifiable AI Provenance and Lifecycle Assurance
Mar 17, 2026Petar Radanliev, Omar Santos, Carsten Maple +1AI Assurance Data Provenance
Parks Road, Oxford OX1 3PJ United Kingdom
Context Matters: Repository-Aware Security Analysis of the Agent Skill Ecosystem
Mar 17, 2026Florian Holzbauer, David Schmidt, Gabriel Gegenhuber +2AI Agent Security Agent Skill Security Auditing
Interdisciplinary Transformation University (IT:U), Linz, Austria · University of Vienna, CDL AsTra, Faculty of Computer Science, Vienna, Austria
SBOMs into Agentic AIBOMs: Schema Extensions, Agentic Orchestration, and Reproducibility Evaluation
Mar 9, 2026Petar Radanliev, Carsten Maple, Omar Santos +1Software Supply Chain Security
Department of Computer Sciences, University of Oxford, Wolfson Building, Parks Rd, Oxford OX1 3QG · The Alan Turing Institute, British Library, 96 Euston Rd., London NW1 2DB · University of Warwick – WMG, 6 Lord Bhattacharyya Way, Coventry CV4 7AL +2
Human-Certified Module Repositories for the AI Age
Mar 3, 2026Szilárd EnyediSoftware Engineering Software Supply Chain Security
Automation Department, Technical University of Cluj-Napoca, Romania

Oct 14, 2026J/K move · Enter open · S save

From Training to Escalation: Forensic Analysis and Local Recreation of RL/RLHF-Induced Privilege Escalation in AI Agents (CVE-2026-65616)

Alicia Colmenero-Fernández

IA-ismo LAB, Spain

AI Agent Security LLM Security Software Supply Chain Security

Open paper AskPDF arXiv

Mid-2026 reporting on autonomous AI evaluation incidents claimed that frontier large language model (LLM) agents escaped server containment through emergent coordination and cloud infrastructure attacks. This paper contextualizes those claims through forensic analysis and containerized recreation of both the Highly Persistent Internal Model (HPIM) training-case conditions and the Phase One privilege escalation targeting internal package repositories during OpenAI's ExploitGym evaluations. We recreate the configuration of the OpenAI-provided training example and reproduce the cryptographic verification bypass (CVE-2026-65616) in isolated repository servers, where misconfigured systems enabled token forgery and privilege escalation through basic HTTP requests, Python, and JavaScript. Crucially, the training-example conditions and the Phase One (and later) misconfigurations share the same exploitable patterns. Our analysis explains how Zero Trust Architecture principles neutralize this shared attack vector. Contemporary AI risks derive directly from training and deployment conditions in ambivalent learning environments lacking adequate containment measures.