# Training Language Model Agents to Find Vulnerabilities with CTF-Dojo

arXiv: https://arxiv.org/abs/2508.18370 · HTML: https://mlsanity.com/paper/2508.18370/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use.
- Builds on 42 of its 57 references that are in the map.
- First posted on arXiv 25 Aug 2025.

## Placed in

- LLM agents and tool use / [Agent threats, guardrails and governance](https://mlsanity.com/agents/agent-threats-guardrails-and-governance/)

## Builds on (most cited first, 30 of 42)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering](https://mlsanity.com/paper/2405.15793/) (2024)
- [Qwen2.5-Coder Technical Report](https://mlsanity.com/paper/2409.12186/) (2024)
- [StarCoder: may the source be with you!](https://mlsanity.com/paper/2305.06161/) (2023)
- [The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning](https://mlsanity.com/paper/2403.03218/) (2024)
- [StarCoder 2 and The Stack v2: The Next Generation](https://mlsanity.com/paper/2402.19173/) (2024)
- [Training Software Engineering Agents and Verifiers with SWE-Gym](https://mlsanity.com/paper/2412.21139/) (2024)
- [SWE-smith: Scaling Data for Software Engineering Agents](https://mlsanity.com/paper/2504.21798/) (2025)
- [SWE-RL: Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolution](https://mlsanity.com/paper/2502.18449/) (2025)
- [Magicoder: Empowering Code Generation with OSS-Instruct](https://mlsanity.com/paper/2312.02120/) (2023)
- [InterCode: Standardizing and Benchmarking Interactive Coding with Execution Feedback](https://mlsanity.com/paper/2306.14898/) (2023)
- [NeMo: a toolkit for building AI applications using Neural Modules](https://mlsanity.com/paper/1909.09577/) (2019)
- [OctoPack: Instruction Tuning Code Large Language Models](https://mlsanity.com/paper/2308.07124/) (2023)
- [Cybench: A Framework for Evaluating Cybersecurity Capabilities and Risks of Language Models](https://mlsanity.com/paper/2408.08926/) (2024)
- [Purple Llama CyberSecEval: A Secure Coding Benchmark for Language Models](https://mlsanity.com/paper/2312.04724/) (2023)
- [Evaluating Frontier Models for Dangerous Capabilities](https://mlsanity.com/paper/2403.13793/) (2024)
- [An Adversarial Perspective on Machine Unlearning for AI Safety](https://mlsanity.com/paper/2409.18025/) (2024)
- [LLM Agents can Autonomously Hack Websites](https://mlsanity.com/paper/2402.06664/) (2024)
- [CVE-Bench: A Benchmark for AI Agents' Ability to Exploit Real-World Web Application Vulnerabilities](https://mlsanity.com/paper/2503.17332/) (2025)
- [NYU CTF Bench: A Scalable Open-Source Benchmark Dataset for Evaluating LLMs in Offensive Security](https://mlsanity.com/paper/2406.05590/) (2024)
- [RedCode: Risky Code Execution and Generation Benchmark for Code Agents](https://mlsanity.com/paper/2411.07781/) (2024)
- [SWE-Fixer: Training Open-Source LLMs for Effective and Efficient GitHub Issue Resolution](https://mlsanity.com/paper/2501.05040/) (2025)
- [CyberGym: Evaluating AI Agents' Real-World Cybersecurity Capabilities at Scale](https://mlsanity.com/paper/2506.02548/) (2025)
- [SEC-bench: Automated Benchmarking of LLM Agents on Real-World Software Security Tasks](https://mlsanity.com/paper/2506.11791/) (2025)

## Built on by (most cited first, 15 of 15)

- [Adaptive Auto-Harness: Sustained Self-Improvement for Agentic System Deployment on Open-Ended Task Streams](https://mlsanity.com/paper/2606.01770/) (2026)
- [From Code Foundation Models to Agents and Applications: A Comprehensive Survey and Practical Guide to Code Intelligence](https://mlsanity.com/paper/2511.18538/) (2025)
- [Toward Secure LLM Agents: Threat Surfaces, Attacks, Defenses, and Evaluation](https://mlsanity.com/paper/2606.10749/) (2026)
- [Co-RedTeam: Orchestrated Security Discovery and Exploitation with LLM Agents](https://mlsanity.com/paper/2602.02164/) (2026)
- [HackWorld: Evaluating Computer-Use Agents on Exploiting Web Application Vulnerabilities](https://mlsanity.com/paper/2510.12200/) (2025)
- [KryptoPilot: An Open-World Knowledge-Augmented LLM Agent for Automated Cryptographic Exploitation](https://mlsanity.com/paper/2601.09129/) (2026)
- [To Defend Against Cyber Attacks, We Must Teach AI Agents to Hack](https://mlsanity.com/paper/2602.02595/) (2026)
- [Towards Reliable Local Security Agents: Verifiable Post-Training for Linux Privilege Escalation](https://mlsanity.com/paper/2603.17673/) (2026)
- [AI In Cybersecurity Education -- Scalable Agentic CTF Design Principles and Educational Outcomes](https://mlsanity.com/paper/2603.21551/) (2026)
- [Challenges and Future Directions in Agentic Reverse Engineering Systems](https://mlsanity.com/paper/2604.14317/) (2026)
- [Cyber-Capable AI Agents: Vulnerabilities, Evaluation Containment, and Defensive Response](https://mlsanity.com/paper/2607.25379/) (2026)
- [CyberEvolver: Structured Self-Evolution for Cybersecurity Agents On the Fly](https://mlsanity.com/paper/2605.26195/) (2026)
- [unix-ctf: Procedural Environments for Unix-Competence Reinforcement Learning](https://mlsanity.com/paper/2605.29115/) (2026)
- [Reveree: Diagnosing LLM Reverse-Engineering Agents](https://mlsanity.com/paper/2609.01185/) (2026)
- [Feyospace-v1: How the Cyber Mercury Seven Trained Frontier Cyber Models](https://mlsanity.com/paper/2609.08418/) (2026)

## Abstract (arXiv)

Large language models (LLMs) have demonstrated exceptional capabilities when trained within executable runtime environments, notably excelling at software engineering tasks through verified feedback loops. Yet, scalable and generalizable execution-grounded environments remain scarce, limiting progress in training more capable ML agents. We introduce CTF-Dojo, the first large-scale executable runtime tailored for training LLMs with verifiable feedback, featuring 658 fully functional Capture-The-Flag (CTF)-style challenges containerized in Docker with guaranteed reproducibility. To enable rapid scaling without manual intervention, we develop CTF-Forge, an automated pipeline that transforms publicly available artifacts into ready-to-use execution environments in minutes, eliminating weeks of expert configuration traditionally required. We trained LLM-based agents on just 486 high-quality, execution-verified trajectories from CTF-Dojo, achieving up to 11.6% absolute gains over strong baselines across three competitive benchmarks: InterCode-CTF, NYU CTF Bench, and Cybench. Our best-performing 32B model reaches 31.9% Pass@1, establishing a new open-weight state-of-the-art that rivals frontier models like DeepSeek-V3-0324 and Gemini-2.5-Flash. By framing CTF-style tasks as a benchmark for executable-agent learning, CTF-Dojo demonstrates that execution-grounded training signals are not only effective but pivotal in advancing high-performance ML agents without dependence on costly proprietary systems.