AgentHound

AgentHound is an open-source offensive security framework designed to assess the security posture of AI agent infrastructure. It performs reconnaissance, fingerprinting, credential looting, and active exploitation across various layers of the agentic stack, mapping attack paths into a Neo4j graph for comprehensive analysis.

AgentHound sits in the Security & Safety category, so it is usually better judged by workflow fit, deployment style, and team needs than by isolated features alone. Based on the available metadata, it leans most clearly toward Full-spectrum attack surface analysis for AI agent infrastructure.. A representative use case is Performing red team engagements against AI systems.. The listed license is Apache-2.0. GitHub stars are currently 452. Relevant tags include Offensive Security, AI Agent Security.