MCP 2026-07-28 spec: stateless core, coming to Claude
Quick Answer
The MCP 2026-07-28 spec introduces a stateless core for Claude, enhancing scalability by allowing AI agents to connect without dedicated sessions.
Quick Take
This shift enables companies to manage thousands of agents efficiently, improving infrastructure resilience and operational simplicity.
Key Points
- 's first version established a standard for AI model connections.
- Stateless MCP simplifies connections, resembling existing web infrastructure.
- Companies can operate thousands of agents making millions of requests.
- Dedicated sessions for each agent hinder scalability and recovery.
- The new spec aims for practical production-scale implementations.
Source Excerpt
The first version of proved that AI models could use a common standard to connect with tools and data.
This version is focused on making those connections practical at production scale.
A company may eventually operate thousands of agents making millions of requests across
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from WebSearch (Tavily)
See more →It's true that Anthropic is the fastest growing company ...
Anthropic is rapidly emerging as a leader in AI with its Claude model, which outperforms competitors in various benchmarks. Their coding agents are also recognized as the best in the industry, contributing to their unprecedented revenue growth and solidifying their position in the enterprise market.
Akshay 🚀 on X: "Karpathy's Agentic Engineering finally has proper tooling! (built by Google) Karpathy defined agentic engineering as the discipline that separates production agent work from vibe coding. The core skills he listed were spec design, eval loops, and security oversight. The https://t.co
Google's new Agents CLI streamlines agentic engineering by integrating seven ADK-specific skills into a single command, enhancing production workflows for coding agents. This tool addresses the fragmented tooling issue in agentic engineering, allowing for seamless scaffolding, evaluation, deployment, and enterprise registration through natural language. Akshay tested this by building a agent from scratch using Claude Code.