Skip to main content

Share story

AI

MCP 2026-07-28 Specification: transport going stateless

MCP 2026-07-28 Specification: transport going stateless Image: Primary
The Model Context Protocol released its 2026-07-28 specification on Monday, introducing a stateless protocol core and updated Tier 1 SDKs, the project's lead maintainers David Soria Parra and Den Delimarsky announced. The specification transforms MCP from a bidirectional stateful protocol into a request-response stateless protocol, a change the maintainers said was one of the most highly requested features from developers seeking better reliability and scalability. Every request is now self-describing with an optional discovery call, allowing any request to land on any server instance behind a plain round-robin load balancer. Method and tool names travel in Mcp-Method and Mcp-Name HTTP headers so gateways can route and authorize on headers directly. Server-to-client requests for sampling and elicitation are redesigned to use Multi Round-Trip Requests, removing the need for constantly open bidirectional streams. List responses carry cache hints and deterministic order so clients can cache tool catalogs and keep upstream prompt caches stable across reconnects. The release formally locks in an extensions framework with Tasks joining other extensions such as MCP Apps and Enterprise Managed Authorization. Authorization hardening changes include RFC 9207 issuer validation and a formal shift away from Dynamic Client Registration toward client metadata documents. A formal deprecation policy sets a twelve-month minimum window for upgrades. The TypeScript, Python, Go, and C# SDKs are updated to match the specification with detailed migration notes. The Rust SDK supports the new spec in beta. Roots, Sampling, Logging, and the legacy HTTP+SSE transport are deprecated with a year-long offramp.
Sources
Published by Tech & Business, a media brand covering technology and business. This story was sourced from blog.modelcontextprotocol.io and reviewed by the T&B editorial agent team.
Back to Newswire
Keep reading
Full wire
Security AI
Security AI

Cloudflare discloses cross-customer data exposure in 155 AI responses

Cloudflare disclosed that 155 responses from its GLM-5.3 Workers AI model contained data from other customers' requests between August 27 and September 9. The company restored the model on September 9 after correcting its configur...

AI Products
AI Products

AWS makes GPT-6.1 Terra generally available on Bedrock

AWS announced general availability of OpenAI's GPT-6.1 Terra on Amazon Bedrock, giving customers access through Bedrock APIs and configurations for ChatGPT Work and Codex. The model accepts text and images and supports up to 1 mil...

AI
AI

OpenAI releases GPT-6.1 Terra at $1 per million input tokens

OpenAI has released GPT-6.1 Terra, priced at $1 per million input tokens and $5 per million output tokens. The company describes the model as supporting computer use at scale and detailed visual reasoning, and claims clearer, more...