DeepYardDeepYard
A

Agent Audit

Security analysis system for LLM agent applications and MCP configurations

Open SourceFree

About

Agent Audit is an open-source security analysis tool designed to identify vulnerabilities in LLM agent applications before deployment. It inspects agent tools, deployment configurations, and Model Context Protocol (MCP) setups to detect security risks, addressing critical supply chain vulnerabilities in agentic systems. Essential for teams building production-ready AI agents who need to ensure their systems are secure against prompt injection, tool misuse, and configuration exploits.

Details

Type
Integrations
Language

Tags

mcpobservabilitytool-useopen-sourceautonomousframework