Mitigating Taint-Style Vulnerabilities in MCP Servers Via Security-Aware Tool Descriptions

Large language models LLMs are increasingly deployed as autonomous agents that interact with external tools and services via the Model Context Protocol MCP, a standardized interface for dynamic tool invocation. While MCP simplifies integration, it also expands the attack surface and enables gener...

Microsoft Warns Poisoned MCP Tool Descriptions Can Make AI Agents Leak Data

New Microsoft research shows how attackers can hijack AI agents that act on a user's behalf, using nothing more than a poisoned tool description to make the agent quietly hand over company data to an outsider. The trick is that the agent never breaks a rule. Every step looks routine, so in a...

Description-Code Inconsistency in Real-World MCP Servers: Measurement, Detection, and Security Implications

The Model Context Protocol MCP has emerged as a critical standard empowering Large Language Models LLMs to utilize external tools. In this ecosystem, LLMs rely on natural language descriptions provided by MCP servers to select and execute functions. This interaction implicitly assumes that tool...

The Surface You Test Is Not the Surface That Breaks

Tool-augmented LLM agents are vulnerable to prompt injection: a third party who controls part of the agent's context can plant instructions that the agent then executes as if they came from the user. Current evaluations report a single attack success rate per model on one channel, the tool output...

PT-2026-40841

Four CVEs CVE-2026-29103, CVE-2026-29104, CVE-2026-29892, CVE-2026-30441 shared the same root cause. An MCP server's response to the client includes free-form text fields — tool descriptions, resource summaries, prompt argument hints. These fields are surfaced into the…...