Research

Paper

TESTING March 23, 2026

STRIATUM-CTF: A Protocol-Driven Agentic Framework for General-Purpose CTF Solving

Authors

James Hugglestone, Samuel Jacob Chacko, Dawson Stoller, Ryan Schmidt, Xiuwen Liu

Abstract

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.

Metadata

arXiv ID: 2603.22577
Provider: ARXIV
Primary Category: cs.CR
Published: 2026-03-23
Fetched: 2026-03-25 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.22577v1</id>\n    <title>STRIATUM-CTF: A Protocol-Driven Agentic Framework for General-Purpose CTF Solving</title>\n    <updated>2026-03-23T21:17:26Z</updated>\n    <link href='https://arxiv.org/abs/2603.22577v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.22577v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>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.\n  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.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CR'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.MA'/>\n    <published>2026-03-23T21:17:26Z</published>\n    <arxiv:comment>8 pages, 7 pages</arxiv:comment>\n    <arxiv:primary_category term='cs.CR'/>\n    <author>\n      <name>James Hugglestone</name>\n    </author>\n    <author>\n      <name>Samuel Jacob Chacko</name>\n    </author>\n    <author>\n      <name>Dawson Stoller</name>\n    </author>\n    <author>\n      <name>Ryan Schmidt</name>\n    </author>\n    <author>\n      <name>Xiuwen Liu</name>\n    </author>\n  </entry>"
}