7 posts tagged with "MCP"

At the start of the year, in AI Security Challenges in 2026, I laid out the challenges the industry as a whole would likely face over the course of the year. Half a year has now passed, and many of the topics I described back then as needing to "wait for future technological maturation" have already reached a stage where we can start taking action. Okta Cross-App Access has become a service ready for production deployment, and Anthropic has adopted it. Sandboxes for agents are becoming standardized, and frontier models are beginning to surpass humans in the application security domain. As we enter the second half of 2026, this article summarizes the latest state of the AI Security industry and lays out where those responsible for AI Security at a typical organization—one that is presumably driving agent adoption—should start over the next six months.

Over the past year, the world of AI has changed a lot. The biggest change is in model performance. One year ago, we were amazed by the release of GPT-4, but looking back now, its performance seems somewhat primitive. The main uses were also limited to rough text generation and summarization.

However, times have moved on. The AI world is shifting from multi-step workflows that simply call LLM APIs to AI agents that run multiple steps autonomously using MCP, tools, and skills. It is now moving further toward a world where multiple AI agents work together on A2A and agent platforms.

In April 2026, Anthropic released Claude Managed Agents as a public beta.

Introducing Claude Managed Agents: everything you need to build and deploy agents at scale. It pairs an agent harness tuned for performance with production infrastructure, so you can go from prototype to launch in days.

— @claudeai

Running AI agents in production requires a lot of work. You need to build an agent loop, set up a sandbox, prepare tool execution infrastructure, and manage credentials. Managed Agents handles all of this as a hosted service from Anthropic, so you can run long-running tasks and async workloads safely without building your own infrastructure.

Since January 2026, OpenClaw has attracted significant attention. Its GitHub star count has surpassed Visual Studio Code, and many forks have appeared, including lightweight versions and reimplementations in other languages. AI agents like OpenClaw that run in the background are being explored for various use cases such as task automation and coding assistance. To keep them running at all times, one click deployment services like exe.new/openclaw and railway.com/deploy/openclaw are now available.

The year 2025 witnessed a continuous cycle of emerging and evolving AI/LLM technologies. Across the industry, various security measures for AI have been advancing. In 2026, AI adoption is expected to expand further, bringing new technologies and demanding corresponding security measures. This article reviews the major topics from 2025, breaks down the anticipated industry-wide challenges in AI Security for 2026 into concrete action items, and summarizes security measures for safely using and providing AI.

MCP, an extension layer for AI and LLM systems, now underpins many AI agents. At the same time, multiple attack vectors have been identified that build on this new AI and LLM context. The best known is MCP tool poisoning, where an agent is tricked into running harmful actions. This was discovered because of the combination of nondeterministic LLM behavior and the new concept of shared context.

While new attack methods emerge, long standing attack vectors are also being exposed by unsafe parts of the MCP specification.

As of July 2025, the Model Context Protocol (MCP) has evolved through three major versions (2024-11-05, 2025-03-26, and 2025-06-18). Today, MCP has become an integral part of our daily development workflow, with developers launching MCP servers via npx commands, uv commands, or Docker containers from MCP clients like Cursor, Claude Code, and VS Code, while also connecting to remote MCP servers.