mlsanityData to 7 Oct 2026

VulnBot: Autonomous Penetration Testing for A Multi-Agent Collaborative Framework

arXiv 2501.13411 ↗First posted 23 Jan 2025cs.SEcited by 30

On the map: LLM agents and tool use → Agent threats, guardrails and governance

Abstract (arXiv)

Penetration testing is a vital practice for identifying and mitigating vulnerabilities in cybersecurity systems, but its manual execution is labor-intensive and time-consuming. Existing large language model (LLM)-assisted or automated penetration testing approaches often suffer from inefficiencies, such as a lack of contextual understanding and excessive, unstructured data generation. This paper presents VulnBot, an automated penetration testing framework that leverages LLMs to simulate the collaborative workflow of human penetration testing teams through a multi-agent system. To address the inefficiencies and reliance on manual intervention in traditional penetration testing methods, VulnBot decomposes complex tasks into three specialized phases: reconnaissance, scanning, and exploitation. These phases are guided by a penetration task graph (PTG) to ensure logical task execution. Key design features include role specialization, penetration path planning, inter-agent communication, and generative penetration behavior. Experimental results demonstrate that VulnBot outperforms baseline models such as GPT-4 and Llama3 in automated penetration testing tasks, particularly showcasing its potential in fully autonomous testing on real-world machines.

Key facts. Cited by 30 papers in the map (data through 7 Oct 2026). Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use. Builds on 25 of its 50 references that are in the map. First posted on arXiv 23 Jan 2025.

Builds on

25 of 50 references.

Closest ancestors · 6

Same direction, newest first

  1. 2024HackSynth: LLM Agent and Evaluation Framework for Autonomous Penetration Testing2 Dec
  2. 2024AutoPenBench: Benchmarking Generative Agents for Penetration Testing4 Oct
  3. 2024BreachSeek: A Multi-Agent Automated Penetration Tester31 Aug
  4. 2024Teams of LLM Agents can Exploit Zero-Day Vulnerabilities2 Jun
  5. 2024LLM Agents can Autonomously Exploit One-day Vulnerabilities11 Apr
  6. 2024Depending on yourself when you should: Mentoring LLM with RL agents to become the master in cybersecurity games26 Mar

Other references · 19

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2023GPT-4 Technical Reportcited by 18,132
  3. 2024The Llama 3 Herd of Modelscited by 14,192
  4. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,500
  5. 2023MetaGPT: Meta Programming for A Multi-Agent Collaborative Frameworkcited by 1,761
  6. 2023Jailbreaking ChatGPT via Prompt Engineering: An Empirical Studycited by 478
  7. 2023MasterKey: Automated Jailbreak Across Multiple Large Language Model Chatbotscited by 172
  8. 2024AutoAttacker: A Large Language Model Guided System to Implement Automatic Cyber-attackscited by 94
  9. 2023Getting pwn'd by AI: Penetration Testing with Large Language Modelscited by 82
  10. 2024CYCLE: Learning to Self-Refine the Code Generationcited by 39
  11. 2024PenHeal: A Two-Stage LLM Framework for Automated Pentesting and Optimal Remediationcited by 30
  12. 2023LLMs as Hackers: Autonomous Linux Privilege Escalation Attackscited by 25
7 more
  1. 2023Exploring ChatGPT's Capabilities on Vulnerability Managementcited by 19
  2. 2024Towards Automated Penetration Testing: Introducing LLM Benchmark, Analysis, and Improvementscited by 11
  3. 2023Understanding Hackers' Work: An Empirical Study of Offensive Security Practitionerscited by 8
  4. 2024CIPHER: Cybersecurity Intelligent Penetration-testing Helper for Ethical Researchercited by 6
  5. 2024Hacking, The Lazy Way: LLM Augmented Pentestingcited by 5
  6. 2024AI-Augmented Ethical Hacking: A Practical Examination of Manual Exploitation and Privilege Escalation in Linux Environmentscited by 3
  7. 2024PTHelper: An open source tool to support the Penetration Testing processcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 16; 2026: 14.

30 papers.

Sort:

By area:

  1. 2025Frontier AI's Impact on the Cybersecurity Landscapecited by 21
  2. 2025Incalmo: An Autonomous LLM-assisted System for Red Teaming Multi-Host Networkscited by 17
  3. 2025A Survey on Agentic Security: Applications, Threats and Defensescited by 17
  4. 2025Benchmarking Practices in LLM-driven Offensive Security: Testbeds, Metrics, and Experiment Designcited by 13
  5. 2026What Makes a Good LLM Agent for Real-world Penetration Testing?cited by 8
  6. 2025PACEbench: A Framework for Evaluating Practical AI Cyber-Exploitation Capabilitiescited by 6
  7. 2025PentestEval: Benchmarking LLM-based Penetration Testing with Modular and Stage-Level Designcited by 6
  8. 2025Pentest-R1: Towards Autonomous Penetration Testing Reasoning Optimized via Two-Stage Reinforcement Learningcited by 4
  9. 2025Hiding in the AI Traffic: Abusing MCP for LLM-Powered Agentic Red Teamingcited by 4
  10. 2025Automated Penetration Testing with LLM Agents and Classical Planningcited by 4
  11. 2025CyberSleuth: Autonomous Blue-Team LLM Agent for Web Attack Forensicscited by 3
  12. 2025Neuro-Symbolic AI for Cybersecurity: State of the Art, Challenges, and Opportunitiescited by 2
18 more
  1. 2025xOffense: An Autonomous Multi-Agent Framework for Penetration Testing with Domain-Adapted Large Language Modelscited by 2
  2. 2026Autonomous LLM Agents & CTFs: A Second Lookcited by 2
  3. 2026AgentSnare: Learning to Delay, Divert, and Defuse Autonomous Penetration Agentscited by 2
  4. 2025Cybersecurity AI: Evaluating Agentic Cybersecurity in Attack/Defense CTFscited by 1
  5. 2026PenForge: On-the-Fly Expert Agent Construction for Automated Penetration Testingcited by 1
  6. 2026A Dual-Loop Agent Framework for Automated Vulnerability Reproductioncited by 1
  7. 2026Red-MIRROR: Agentic LLM-based Autonomous Penetration Testing with Reflective Verification and Knowledge-augmented Interactioncited by 1
  8. 2026The Ethics of Autonomous AI Agents for Offensive Securitycited by 1
  9. 2025Recognition Without Mitigation: Ethical Frameworks in Autonomous Offensive-LLM Agent Researchcited by 0
  10. 2025Breaking Android with AI: A Deep Dive into LLM-Powered Exploitationcited by 0
  11. 2025SoK: Measuring What Matters for Closed-Loop Security Agentscited by 0
  12. 2026AutoEG: Exploiting Known Third-Party Vulnerabilities in Black-Box Web Applicationscited by 0
  13. 2026Challenges and Future Directions in Agentic Reverse Engineering Systemscited by 0
  14. 2026Pen-Strategist: A Reasoning Framework for Penetration Testing Strategy Formation and Analysiscited by 0
  15. 2026From Controlled to the Wild: Evaluation of Pentesting Agents for the Real-Worldcited by 0
  16. 2026CyberEvolver: Structured Self-Evolution for Cybersecurity Agents On the Flycited by 0
  17. 2026Staying on the Attack Path: Structured State for Long-Horizon Automated Penetration Testingcited by 0
  18. 2026Toward Secure AI-Powered Penetration Testing Agents: Security Threats, Guardrails, and Architectural Perspectivescited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Vulnerability Mitigation System (VMS): LLM Agent and Evaluation Framework for Autonomous Penetration Testing
  2. 2024PentestAgent: Incorporating LLM Agents to Automated Penetration Testing
  3. 2026Environment-Grounded Multi-Agent Workflow for Autonomous Penetration Testing
  4. 2025AutoPentester: An LLM Agent-based Framework for Automated Pentesting
  5. 2025From Capabilities to Performance: Evaluating Key Functional Properties of LLM Architectures in Penetration Testing
  6. 2026Hackers or Hallucinators? A Comprehensive Analysis of LLM-Based Automated Penetration Testing
  7. 2025Construction and Evaluation of LLM-based agents for Semi-Autonomous penetration testing
  8. 2025AutoPentest: Enhancing Vulnerability Management With Autonomous LLM Agents