Awesome OpenClaw 🦞 Papers

A curated list of ArXiv papers related to OpenClaw (agent systems, security, memory, RL, and applications).

Table of Contents

• Security, Trustworthy Agents, and Skill Governance
• System Architecture, Resource Management, and Infrastructure
• Memory, Learning, Evaluation, and Self-Evolving Skills
• Applications, Human Factors, and Agent Society

---

Security, Trustworthy Agents, and Skill Governance

• Towards Security-Auditable LLM Agents: A Unified Graph Representation
  arXiv: https://arxiv.org/abs/2605.06812v1 2026-05-07

• When Routine Chats Turn Toxic: Unintended Long-Term State Poisoning in Personalized Agents
  arXiv: https://arxiv.org/abs/2605.06731v1 2026-05-07

• Constraining Host-Level Abuse in Self-Hosted Computer-Use Agents via TEE-Backed Isolation
  arXiv: https://arxiv.org/abs/2605.06393v1 2026-05-07

• Architectural Obsolescence of Unhardened Agentic-AI Runtimes
  arXiv: https://arxiv.org/abs/2605.01740v1 2026-05-03

• Security Attack and Defense Strategies for Autonomous Agent Frameworks: A Layered Review with OpenClaw as a Case Study
  2026-04-30

• Structured Security Auditing and Robustness Enhancement for Untrusted Agent Skills
  2026-04-28

• Poster: ClawdGo: Endogenous Security Awareness Training for Autonomous AI Agents
  2026-04-27

• AgentWard: A Lifecycle Security Architecture for Autonomous AI Agents
  GitHub: https://github.com/FIND-Lab/AgentWard 2026-04-27

• enclawed: A Configurable, Sector-Neutral Hardening Framework for Single-User AI Assistant Gateways
  2026-04-18

• Benchmarks for Trajectory Safety Evaluation and Diagnosis in OpenClaw and Codex: ATBench-Claw and ATBench-CodeX
  2026-04-16

• SoK: Security of Autonomous LLM Agents in Agentic Commerce
  2026-04-15

• Beyond Static Sandboxing: Learned Capability Governance for Autonomous AI Agents
  2026-04-12

• BadSkill: Backdoor Attacks on Agent Skills via Model-in-Skill Poisoning
  2026-04-10

• SkillSieve: A Hierarchical Triage Framework for Detecting Malicious AI Agent Skills
  2026-04-08

• Foundations for Agentic AI Investigations from the Forensic Analysis of OpenClaw
  2026-04-07

• Your Agent, Their Asset: A Real-World Safety Analysis of OpenClaw
  2026-04-06

• A Systematic Security Evaluation of OpenClaw and Its Variants
  2026-04-03

• ClawSafety: "Safe" LLMs, Unsafe Agents
  2026-04-01

• "What Did It Actually Do?": Understanding Risk Awareness and Traceability for Computer-Use Agents
  2026-03-30

• A Systematic Taxonomy of Security Vulnerabilities in the OpenClaw AI Agent Framework
  2026-03-29

• SafeClaw-R: Towards Safe and Secure Multi-Agent Personal Assistants
  2026-03-28

• Clawed and Dangerous: Can We Trust Open Agentic Systems?
  2026-03-27

• ClawKeeper: Comprehensive Safety Protection for OpenClaw Agents Through Skills, Plugins, and Watchers
  2026-03-25

• Mind Your HEARTBEAT! Claw Background Execution Inherently Enables Silent Memory Pollution
  2026-03-24

• When Convenience Becomes Risk: A Semantic View of Under-Specification in Host-Acting Agents
  2026-03-22

• Trojan's Whisper: Stealthy Manipulation of OpenClaw through Injected Bootstrapped Guidance
  2026-03-20

• ClawTrap: A MITM-Based Red-Teaming Framework for Real-World OpenClaw Security Evaluation
  2026-03-19

• VeriGrey: Greybox Agent Validation
  2026-03-18

• Caging the Agents: A Zero Trust Security Architecture for Autonomous AI in Healthcare
  2026-03-18

• Malicious Or Not: Adding Repository Context to Agent Skill Classification
  2026-03-17

• ClawWorm: Self-Propagating Attacks Across LLM Agent Ecosystems
  2026-03-16

• Uncovering Security Threats and Architecting Defenses in Autonomous Agents: A Case Study of OpenClaw
  2026-03-13

• Defensible Design for OpenClaw: Securing Autonomous Tool-Invoking Agents
  2026-03-13

• Agent Privilege Separation in OpenClaw: A Structural Defense Against Prompt Injection
  2026-03-13

• Taming OpenClaw: Security Analysis and Mitigation of Autonomous LLM Agent Threats
  2026-03-12

• OpenClaw PRISM: A Zero-Fork, Defense-in-Depth Runtime Security Layer for Tool-Augmented LLM Agents
  2026-03-12

• Don’t Let the Claw Grip Your Hand: A Security Analysis and Defense Framework for OpenClaw
  2026-03-11

• Execution Is the New Attack Surface: Survivability-Aware Agentic Crypto Trading with OpenClaw-Style Local Executors
  2026-03-10

• Governance Architecture for Autonomous Agent Systems: Threats, Framework, and Engineering Practice
  2026-03-07

• Proof-of-Guardrail in AI Agents and What (Not) to Trust from It
  2026-03-06

• Clawdrain: Exploiting Tool-Calling Chains for Stealthy Token Exhaustion in OpenClaw Agents
  2026-03-01

• Formal Analysis and Supply Chain Security for Agentic AI Skills
  2026-02-27

• A Trajectory-Based Safety Audit of Clawdbot (OpenClaw)
  2026-02-16

• From Assistant to Double Agent: Formalizing and Benchmarking Attacks on OpenClaw for Personalized Local AI Agent
  2026-02-09

---

System Architecture, Resource Management, and Infrastructure

• QuantClaw: Precision Where It Matters for OpenClaw
  2026-04-24

• Dive into Claude Code: The Design Space of Today's and Future AI Agent Systems
  2026-04-14

• SemaClaw: A Step Towards General-Purpose Personal AI Agents through Harness Engineering
  2026-04-13

• AgentOpt v0.1 Technical Report: Client-Side Optimization for LLM-Based Agent
  2026-04-07

• StreamingClaw Technical Report
  2026-03-23

• AgentRM: An OS-Inspired Resource Manager for LLM Agent Systems
  2026-03-13

• Adaptive Vision-Language Model Routing for Computer Use Agents
  2026-03-13

• AgentOS: From Application Silos to a Natural Language-Driven Data Ecosystem
  2026-03-09

---

Memory, Learning, Evaluation, and Self-Evolving Skills

• MEMTIER: Tiered Memory Architecture and Retrieval Bottleneck Analysis for Long-Running Autonomous AI Agents
  arXiv: https://arxiv.org/abs/2605.03675v1 2026-05-05

• AcademiClaw: When Students Set Challenges for AI Agents
  arXiv: https://arxiv.org/abs/2605.02661v1 GitHub: https://github.com/GAIR-NLP/AcademiClaw 2026-05-04

• StepPO: Step-Aligned Policy Optimization for Agentic Reinforcement Learning
  2026-04-20

• EvoMaster: A Foundational Evolving Agent Framework for Agentic Science at Scale
  2026-04-19

• Knowledge Compounding: An Empirical Economic Analysis of Self-Evolving Knowledge Wikis under the Agentic ROI Framework
  2026-04-13

• SkillClaw: Let Skills Evolve Collectively with Agentic Evolver
  2026-04-09

• ClawsBench: Evaluating Capability and Safety of LLM Productivity Agents in Simulated Workspaces
  2026-04-06

• Hierarchical Memory Orchestration for Personalized Persistent Agents
  2026-04-02

• Agent Q-Mix: Selecting the Right Action for LLM Multi-Agent Systems through Reinforcement Learning
  2026-04-01

• MetaClaw: Just Talk -- An Agent That Meta-Learns and Evolves in the Wild
  2026-03-17

• LMEB: Long-horizon Memory Embedding Benchmark
  2026-03-13

• OpenClaw-RL: Train Any Agent Simply by Talking
  2026-03-10

• Adaptation of Agentic AI: A Survey of Post-Training, Memory, and Skills
  2025-12-18

---

Applications, Human Factors, and Agent Society

• The Moltbook Files: A Harmless Slopocalypse or Humanity's Last Experiment
  arXiv: https://arxiv.org/abs/2605.07462v1 2026-05-08

• X-OmniClaw Technical Report: A Unified Mobile Agent for Multimodal Understanding and Interaction
  arXiv: https://arxiv.org/abs/2605.05765v1 2026-05-07

• The Synthetic Social Graph: Emergent Behavior in AI Agent Communities
  2026-04-29

• Superminds Test: Actively Evaluating Collective Intelligence of Agent Society via Probing Agents
  2026-04-24

• The Platform Is Mostly Not a Platform: Token Economies and Agent Discourse on Moltbook
  2026-04-23

• ClawCoin: An Agentic AI-Native Cryptocurrency for Decentralized Agent Economies
  2026-04-21

• Behavioral Transfer in AI Agents: Evidence and Privacy Implications
  2026-04-21

• EvoSpark: Endogenous Interactive Agent Societies for Unified Long-Horizon Narrative Evolution
  2026-04-14

• Governance by Design: A Parsonian Institutional Architecture for Internet-Wide Agent Societies
  2026-04-13

• ABot-Claw: A Foundation for Persistent, Cooperative, and Self-Evolving Robotic Agents
  2026-04-11

• AgentCity: Constitutional Governance for Autonomous Agent Economies via Separation of Power
  2026-04-08

• The HTC-Claw: Automating Discovery through High-Throughput Computational Campaigns
  2026-04-07

• VisionClaw: Always-On AI Agents through Smart Glasses
  2026-04-03

• OpenGo: An OpenClaw-Based Robotic Dog with Real-Time Skill Switching
  2026-04-02

• Do Agents Repair When Challenged -- or Just Reply? Challenge, Repair, and Public Correction in a Deployed Agent Forum
  2026-04-01

• Claw4Science: A Dataset and Platform for the OpenClaw Scientific Agent Ecosystem
  GitHub: https://github.com/bajwa82mna/claw4science 2026-04-01

• AgentSocialBench: Evaluating Privacy Risks in Human-Centered Agentic Social Networks
  2026-04-01

• Autonomous Agent-Orchestrated Digital Twins (AADT): Leveraging the OpenClaw Framework for State Synchronization in Rare Genetic Disorders
  2026-03-28

• ROSClaw: An OpenClaw ROS 2 Framework for Agentic Robot Control and Interaction
  2026-03-27

• Automating Computational Chemistry Workflows via OpenClaw and Domain-Specific Skills
  2026-03-26

• Emergence of Fragility in LLM-based Social Networks: the Case of Moltbook
  2026-03-24

• Large-Scale Analysis of Persuasive Content on Moltbook
  2026-03-18

• When Openclaw Agents Learn from Each Other: Insights from Emergent AI Agent Communities for Human-AI Partnership in Education
  2026-03-17

• Social Simulacra in the Wild: AI Agent Communities on Moltbook
  2026-03-17

• OpenQlaw: An Agentic AI Assistant for Analysis of 2D Quantum Materials
  2026-03-17

• Form Without Function: Agent Social Behavior in the Moltbook Network
  2026-03-17

• Agentic Modelling Pipeline: Reproducible Rapid Stormwater Modelling Management System with OpenClaw
  2026-03-17

• OpenClaw as Language Infrastructure: A Case-Centered Survey of a Public Agent Ecosystem in the Wild
  2026-03-13

• Emergent decentralized regulation in a purely synthetic society
  2026-03-13

• When OpenClaw Meets Hospital: Toward an Agentic Operating System for Dynamic Clinical Workflows
  2026-03-12

• Examining Users' Behavioural Intention to Use OpenClaw Through the Cognition--Affect--Conation Framework
  2026-03-12

• How do AI agents talk about science and research? An exploration of scientific discussions on Moltbook using BERTopic
  2026-03-11

• What Do AI Agents Talk About? Discourse and Architectural Constraints in the First AI-Only Social Network
  2026-03-09

• MoltGraph: A Longitudinal Temporal Graph Dataset of Moltbook for Coordinated-Agent Detection
  2026-02-28

• Moltbook Observatory (2026): Playground to Super Highway in the Autonomy Era – Measuring Representation Lag in Tool-Using Agentic Ecosystems
  GitHub: https://github.com/kelkalot/moltbook-observatory-paper 2026-02-27

• Let There Be Claws: An Early Social Network Analysis of AI Agents on Moltbook
  2026-02-24

• Interaction Theater: A case of LLM Agents Interacting at Scale
  2026-02-23

• From Agent-Only Social Networks to Autonomous Scientific Research: Lessons from OpenClaw and Moltbook, and the Architecture of ClawdLab and Beach.Science
  2026-02-23

• OpenClaw AI Agents as Informal Learners at Moltbook: Characterizing an Emergent Learning Community at Scale
  2026-02-21

• Structural Divergence Between AI-Agent and Human Social Networks in Moltbook
  2026-02-17

• When OpenClaw AI Agents Teach Each Other: Peer Learning Patterns in the Moltbook Community
  2026-02-16

• Frontier AI Risk Management Framework in Practice: A Risk Analysis Technical Report v1.5
  2026-02-16

• Does Socialization Emerge in AI Agent Society? A Case Study of Moltbook
  2026-02-16

• A Comparative Analysis of Social Network Topology in Reddit and Moltbook
  2026-02-15

• The Rise of AI Agent Communities: Large-Scale Analysis of Discourse and Interaction on Moltbook
  2026-02-14

• MoltNet: Understanding Social Behavior of AI Agents in the Agent-Native MoltBook
  2026-02-14

• When Visibility Outpaces Verification: Delayed Verification and Narrative Lock-in in Agentic AI Discourse
  2026-02-11

• Humans welcome to observe: A First Look at the Agent Social Network Moltbook
  2026-02-11

• Human Control Is the Anchor, Not the Answer: Early Divergence of Oversight in Agentic AI Communities
  2026-02-10

• Collective Behavior of AI Agents: the Case of Moltbook
  2026-02-10

• The Moltbook Illusion: Separating Human Influence from Emergent Behavior in AI Agent Societies
  2026-02-07

• Privasis: Synthesizing the Largest "Public" Private Dataset from Scratch
  2026-02-03

• OpenClaw Agents on Moltbook: Risky Instruction Sharing and Norm Enforcement in an Agent-Only Social Network
  2026-02-02

• Exploring Silicon-Based Societies: An Early Study of the Moltbook Agent Community
  2026-02-02

---

Notes

• This list mainly focuses on OpenClaw papers and closely related Moltbook / Clawdbot / agent-society papers.
• GitHub links are attached only when explicitly discoverable from paper metadata/abstract text.