Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.22577 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908908690341888 |
|---|---|
| author | Hugglestone, James Chacko, Samuel Jacob Stoller, Dawson Schmidt, Ryan Liu, Xiuwen |
| author_facet | Hugglestone, James Chacko, Samuel Jacob Stoller, Dawson Schmidt, Ryan Liu, Xiuwen |
| contents | Large Language Models (LLMs) have demonstrated potential in code generation, yet they struggle with the multi-step, stateful reasoning required for offensive cybersecurity operations. Existing research often relies on static benchmarks that fail to capture the dynamic nature of real-world vulnerabilities. In this work, we introduce STRIATUM-CTF (A Search-based Test-time Reasoning Inference Agent for Tactical Utility Maximization in Cybersecurity), a modular agentic framework built upon the Model Context Protocol (MCP). By standardizing tool interfaces for system introspection, decompilation, and runtime debugging, STRIATUM-CTF enables the agent to maintain a coherent context window across extended exploit trajectories.
We validate this approach not merely on synthetic datasets, but in a live competitive environment. Our system participated in a university-hosted Capture-the-Flag (CTF) competition in late 2025, where it operated autonomously to identify and exploit vulnerabilities in real-time. STRIATUM-CTF secured First Place, outperforming 21 human teams and demonstrating strong adaptability in a dynamic problem-solving setting. We analyze the agent's decision-making logs to show how MCP-based tool abstraction significantly reduces hallucination compared to naive prompting strategies. These results suggest that standardized context protocols are a critical path toward robust autonomous cyber-reasoning systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_22577 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | STRIATUM-CTF: A Protocol-Driven Agentic Framework for General-Purpose CTF Solving Hugglestone, James Chacko, Samuel Jacob Stoller, Dawson Schmidt, Ryan Liu, Xiuwen Cryptography and Security Artificial Intelligence Multiagent Systems I.2.1 Large Language Models (LLMs) have demonstrated potential in code generation, yet they struggle with the multi-step, stateful reasoning required for offensive cybersecurity operations. Existing research often relies on static benchmarks that fail to capture the dynamic nature of real-world vulnerabilities. In this work, we introduce STRIATUM-CTF (A Search-based Test-time Reasoning Inference Agent for Tactical Utility Maximization in Cybersecurity), a modular agentic framework built upon the Model Context Protocol (MCP). By standardizing tool interfaces for system introspection, decompilation, and runtime debugging, STRIATUM-CTF enables the agent to maintain a coherent context window across extended exploit trajectories. We validate this approach not merely on synthetic datasets, but in a live competitive environment. Our system participated in a university-hosted Capture-the-Flag (CTF) competition in late 2025, where it operated autonomously to identify and exploit vulnerabilities in real-time. STRIATUM-CTF secured First Place, outperforming 21 human teams and demonstrating strong adaptability in a dynamic problem-solving setting. We analyze the agent's decision-making logs to show how MCP-based tool abstraction significantly reduces hallucination compared to naive prompting strategies. These results suggest that standardized context protocols are a critical path toward robust autonomous cyber-reasoning systems. |
| title | STRIATUM-CTF: A Protocol-Driven Agentic Framework for General-Purpose CTF Solving |
| topic | Cryptography and Security Artificial Intelligence Multiagent Systems I.2.1 |
| url | https://arxiv.org/abs/2603.22577 |