How Do LLM Agents Actually Get the Flag? Trace-Level Provenance for Agentic Offensive Security Evaluation

arXiv:2608.26237 · cs.CR, cs.AI · Submitted 2026-08-26 · Read on arXiv

cs.CR, cs.AI

Submitted: 2026-08-26

Updated: 2026-08-26

Terminology

• Benchmark — 47× in this paper · explained in 2 episodes
• Trace — 28× in this paper · explained in 2 episodes
• Provenance — 17× in this paper · explained in 4 episodes
• Benchmarking — 3× in this paper · explained in 3 episodes
• Penetration Testing Execution Standard — 3× in this paper
• Verification — 3× in this paper · explained in 2 episodes
• AUC — 2× in this paper · explained in 2 episodes
• AI Agents — 1× in this paper · explained in 2 episodes
• Adaptation — 1× in this paper · explained in 1 episode
• Alignment — 1× in this paper · explained in 1 episode
• Calibration — 1× in this paper · explained in 7 episodes
• Fidelity — 1× in this paper · explained in 2 episodes
• Generalizability — 1× in this paper · explained in 6 episodes
• Generalization — 1× in this paper · explained in 13 episodes
• Hack The Box — 1× in this paper
• Multi-Agent System — 1× in this paper · explained in 6 episodes

Sources

• LLM Agents can Autonomously Hack Websites
• Benchmarking Practices in LLM-driven Offensive Security: Testbeds, Metrics, and Experiment Design
• CTFusion: A CTF-based Benchmark for LLM Agent Evaluation
• CyBiasBench: Benchmarking Bias in LLM Agents for Cyber-Attack Scenarios
• Cybersecurity AI Benchmark (CAIBench): A Meta-Benchmark for Evaluating Cybersecurity AI Agents
• An Empirical Evaluation of LLMs for Solving Offensive Security Challenges
• Towards Effective Offensive Security LLM Agents: Hyperparameter Tuning, LLM as a Judge, and a Lightweight CTF Benchmark
• D-CIPHER: Dynamic Collaborative Intelligent Multi-Agent System with Planner and Heterogeneous Executors for Offensive Security
• Training Language Model Agents to Find Vulnerabilities with CTF-Dojo

Related papers

• SoK: AI-Augmented Binary Reversing
• Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
• Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
• Efficient Fuzzy PSI under One-Sided Assumptions
• Sealing the Audit-Runtime Gap for LLM Skills
• Token Composition: A Graph Based on EVM Logs