mcp news, stories and analysis

Donut chart showing that 58.6 percent of agentic MCP exploit sequences bypass SOTA guardrails, while only 41.4 percent are caught. Data from the Agentic Safety Evaluation dataset of 200 CVE-derived tool-call chains tested against five production guardrails.

Agentic MCP attacks bypass SOTA guardrails 58% of the time

MCP attack chains bypass SOTA guardrails more than half the time because text classifiers miss composed tool-call exploits. The agentic safety gap is architectural, not a tuning problem.

Lars Cornelissen · Jul 9, 2026

Abstract data-art visualization contrasting direct API integrations that scale as apps times tools with MCP connectors that scale as apps plus tools

AI for dummies

MCP explained: what it actually adds beyond a normal API

MCP, the Model Context Protocol, is an open standard that lets AI models discover and call tools at runtime instead of you hardcoding an API for each one.

Lars Cornelissen · Jul 6, 2026

Spring AI 2.0 release work by GitHub release-note count: 5 new features, 5 bug fixes, 4 documentation updates, and 2 dependency upgrades.

Engineering

Spring AI 2.0 gives Java agents a sturdier app stack

Spring AI 2.0 is a sturdier Java AI stack: Boot 4.1, MCP, unified tool loops, and 4 Cosmos DB modules now matter for agents.

Lars Cornelissen · Jun 30, 2026

AutoJack exposure chart showing one malicious page, three chained weaknesses, zero stable PyPI releases affected, and one GitHub hardening commit.

Cyber security

AutoJack makes AI agent prototypes a real RCE risk

AutoJack is a host RCE warning for AI agent prototypes: one malicious page chained three AutoGen Studio weaknesses into command execution.

Lars Cornelissen · Jun 23, 2026

Line chart showing bespoke connectors needed grows to 50 without MCP for 5 clients and 10 tools, versus only 15 with MCP, because the cost goes from clients times tools to clients plus tools.

Engineering

What is MCP? The Model Context Protocol for data engineers

MCP is an open standard that lets AI models call your tools and data through one connector instead of a custom integration per model. Here is what it means for data engineers.

Lars Cornelissen · Jun 10, 2026