STRIATUM-CTF: A Protocol-Driven Agentic Framework for General-Purpose CTF Solving
This addresses the problem of autonomous cybersecurity operations for researchers and practitioners, offering a novel framework with demonstrated competitive success.
The paper tackles the challenge of LLMs struggling with multi-step reasoning in cybersecurity by introducing STRIATUM-CTF, a modular agentic framework that uses standardized tool interfaces to maintain context across exploit trajectories. The system achieved First Place in a university-hosted CTF competition, outperforming 21 human teams and demonstrating strong adaptability in real-time problem-solving.
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.