Research library
Curate the field by topic, not just recency.
This section is the long-lived knowledge layer for important papers, organizing the multi-agent security literature into durable subareas that can support reading lists, annotated references, and future topic pages.
Stored article buckets
These groups come from the categorized article database and show a preview of the latest papers in each bucket.
925 papers
Governance and Policy
- Cybersecurity in Autonomous AI Robotics: A Review of Emerging Threats, Adversarial Attacks, and Mitigation Techniques
- The Governance Gap in Agentic Memory
- Kill-Switch Doctrine Gap in Gulf Sovereign AI Infrastructure
599 papers
Benchmarks and Evaluation
- Cybersecurity in Autonomous AI Robotics: A Review of Emerging Threats, Adversarial Attacks, and Mitigation Techniques
- Beyond Injection Detection: A Positive-Security Prompt Firewall that Closes the Scope and PHI Gap SOTA Classifiers Miss in Healthcare
- Token Budgets: Replication Package
560 papers
Orchestration Risk
- Cybersecurity in Autonomous AI Robotics: A Review of Emerging Threats, Adversarial Attacks, and Mitigation Techniques
- Kill-Switch Doctrine Gap in Gulf Sovereign AI Infrastructure
- Token Budgets: Replication Package
532 papers
Agent-to-Agent Communication
- An LLM Agent Cannot Be a Gate: Why a Recited Rule Is Not an Enforced One
- Beyond tokens: a unified framework for latent communication in LLM-based multi-agent systems
- LLM-Guided Digital Twin Agents for Autonomous Threat Detection and Response in Cyber-Physical Energy Systems
497 papers
Trust and Identity
- Beyond Injection Detection: A Positive-Security Prompt Firewall that Closes the Scope and PHI Gap SOTA Classifiers Miss in Healthcare
- The Governance Gap in Agentic Memory
- AI-Driven Network Security in Next-Generation 5G/6G Smart Environments
289 papers
Prompt Injection
- Beyond Injection Detection: A Positive-Security Prompt Firewall that Closes the Scope and PHI Gap SOTA Classifiers Miss in Healthcare
- Agent Infrastructure Engineer: The New DevOps
- Membrane: A Self-Evolving Contrastive Safety Memory for LLM Agent Defense
135 papers
Memory Poisoning
- WebMCP Tool Surface Poisoning: Runtime Manipulation Attacks on LLM Agents
- Beyond Similarity: Trustworthy Memory Search for Personal AI Agents
- Membrane: A Self-Evolving Contrastive Safety Memory for LLM Agent Defense