Skip to content

Commit 7f48c5a

Browse files
authored
Updated draft for MCP Spec Timeline (modelcontextprotocol#1550)
Updated draft blog for MCP timeline. Worked in collab with David SP and Adam J.
1 parent ce508ef commit 7f48c5a

File tree

1 file changed

+30
-34
lines changed

1 file changed

+30
-34
lines changed
Lines changed: 30 additions & 34 deletions
Original file line numberDiff line numberDiff line change
@@ -1,43 +1,39 @@
11
---
2-
title: "Update on the Next MCP Protocol Release"
3-
date: 2025-09-21T10:00:00-08:00
2+
title: "Update on the Next MCP Spec Release"
3+
date: 2025-09-26T10:00:00-08:00
44
draft: true
55
tags: ["mcp", "protocol", "roadmap", "community"]
66
author: "David Soria Parra"
77
description: "An update on the timeline and priorities for the next Model Context Protocol specification version"
88
---
99

10-
## Release Timeline Update
10+
The next version of the Model Context Protocol will be released on **November 25th, 2025**. This timeline allows us to focus on delivering important improvements needed for the protocol's evolution, such as asynchronous tasks, stateless transport enhancements, and server identity, while ensuring our [new governance model](https://modelcontextprotocol.io/community/governance) is functioning effectively.
1111

12-
The next version of the Model Context Protocol specification will be released on **November 25th, 2025**, with a release candidate available on **November 11th, 2025**. This 14-day release candidate period will allow client implementors and SDK maintainers to validate protocol changes before the final release. This updated timeline allows for proper focus on delivering critical improvements needed for the protocol's evolution and ensures our [new governance model](https://modelcontextprotocol.io/community/governance) is functioning effectively.
12+
## Progress Since Last Spec Release
1313

14-
## Summer Progress
14+
Our last spec was released on June 18, 2025, and focused on structured tool outputs, OAuth-based authorization, elicitation for server-initiated user interactions, and improved security best practices.
1515

16-
Over the past few months, we've focused on establishing the foundations for the broader MCP ecosystem:
16+
Since then, weve focused on establishing additional foundations for the MCP ecosystem:
1717

1818
### Formal governance structures
1919

20-
We established a [formal governance model for MCP](https://modelcontextprotocol.io/community/governance), including defined roles and decision-making mechanisms. Additionally, we developed the Specification Enhancement Proposal (SEP) process to provide clear guidelines for contributing specification changes and transparent decision-making procedures.
21-
22-
Our governance model is still in its early stages and we're iterating on making it scale better as the protocol and the community around it grows.
20+
We established a [formal governance model](https://modelcontextprotocol.io/community/governance) for MCP, including defined roles and decision-making mechanisms. Additionally, we developed the [Specification Enhancement Proposal (SEP)](https://modelcontextprotocol.io/community/sep-guidelines) process to provide clear guidelines for contributing specification changes and transparent decision-making procedures. This process is still in its early phase and we’re working on making it smoother and faster.
2321

2422
### Working groups
2523

26-
We've established [Working Groups and Interest Groups](https://modelcontextprotocol.io/community/working-interest-groups) to foster community collaboration and drive the specification forward. These groups serve multiple purposes: they provide clear entry points for new contributors, empower community members to lead initiatives in their areas of expertise, and distribute ownership across the ecosystem rather than concentrating it among core maintainers.
27-
28-
While the working group model is evolving, we're also actively developing governance structures that will grant these groups greater autonomy in decision-making and implementation. This distributed approach ensures the protocol can grow sustainably while maintaining quality and consistency across different focus areas.
29-
30-
### Registry development
24+
To foster community collaboration and drive the specification forward, we established [Working and Interest Groups](https://modelcontextprotocol.io/community/working-interest-groups). These groups serve multiple purposes:
3125

32-
We [launched the MCP Registry in preview](https://blog.modelcontextprotocol.io/posts/2025-09-08-mcp-registry-preview/) as an open catalog and API for publicly available MCP servers to improve discoverability across the ecosystem. This centralized infrastructure serves as a single source of truth for MCP servers, supporting both public and private sub-registries that organizations can customize for their specific needs.
26+
1. Provide clear entry points for new contributors
27+
2. Empower community members to lead initiatives in their areas of expertise
28+
3. Distribute ownership across the ecosystem rather than concentrating it among core maintainers
3329

34-
The MCP registry represents a collaborative effort involving multiple companies and developers, with community-driven moderation mechanisms to maintain a high quality bar.
30+
While our working group model is still evolving, we're actively developing governance structures that will grant these groups greater autonomy in decision-making and implementation. This distributed approach ensures the protocol can grow sustainably while maintaining quality and consistency across different focus areas.
3531

36-
MCP clients can also consume registry content via registry aggregators, enabling users to easily discover and integrate MCP servers into their AI workflows.
32+
### Registry development
3733

38-
All of these initiatives set the foundation for MCP's continued development. We know that there is more work ahead, and we want to lean on the broad expertise of our community to help shape the protocol in a way that serves it.
34+
To improve discoverability across the ecosystem, we launched the [MCP Registry](https://blog.modelcontextprotocol.io/posts/2025-09-08-mcp-registry-preview/) in preview as an open catalog and API for publicly available MCP servers. This centralized infrastructure serves as a single location where clients can discover MCP servers they want to use. It’s backed by a standardized API spec, supporting both public and private sub-registries that organizations can customize for their specific needs.
3935

40-
Our governance model and formalized working groups will enable focused development on the highest priority protocol improvements while incorporating community input and contributions. MCP continues to be an **open protocol**, where **anyone can contribute**.
36+
These initiatives provide the foundation for MCP's continued development. The established governance model and working groups enable focused development on high-priority protocol improvements, while the registry provides an open and community-driven approach to expanding server adoption. We’re excited to see these initiatives grow in the coming months as we focus on the next specification release.
4137

4238
## Priority Areas for the Next Release
4339

@@ -47,43 +43,43 @@ Implementing support for asynchronous tasks and tool calling in MCP to enable lo
4743

4844
### Statelessness and Scalability
4945

50-
Improving the protocol's stateless capabilities to enable easier scaling for MCP server deployments in large production environments.
46+
Improving the protocol's statelessness capabilities to enable easier scaling for MCP server deployments in large production environments.
5147

52-
While Streamable HTTP supports stateless deployments of MCP servers, production environments face challenges with initialization sequences and session management. The Transport Working Group is developing improvements to the transport layer that will support stateless MCP while maintaining straightforward upgrade paths to stateful implementations when needed.
48+
While Streamable HTTP supports stateless deployments of MCP servers, production environments face challenges with initialization sequences and session management. These issues complicate real-world stateless deployments using Streamable HTTP. The **Transport Working Group** is developing improvements to the transport layer that will support stateless MCP while maintaining straightforward upgrade paths to stateful implementations when needed
5349

5450
### Server Identity
5551

56-
Implementing server identity through well-known URLs that expose server metadata, improving discovery and authorization mechanisms.
52+
Implementing server identity through well-known URLs that expose server metadata, improving discovery and authentication mechanisms.
5753

58-
Currently, clients must initialize a connection to an MCP server to obtain server information. This requirement complicates discovery for clients and crawlers (such as registry systems). The planned implementation will use the standardized [`.well-known` format](https://en.wikipedia.org/wiki/Well-known_URI), allowing server authors to expose MCP server information in a static, cacheable, and easily discoverable manner.
54+
Currently, clients must initialize a connection to an MCP server to obtain server information, which complicates discovery for clients and crawlers (such as registry systems). The planned implementation will use the standardized .well-known format, allowing server authors to expose MCP server information in a static, cacheable, and easily discoverable manner.
5955

6056
### Official Extensions
6157

62-
Establishing officially endorsed extensions for MCP. We see patterns emerging in the MCP ecosystem that are highly relevant to specific use cases and industry domains. While not all of them can be embedded directly into the specification, we recognize that there is still a very real need to provide extensibility points that enable folks to build the experiences they need on top of MCP.
63-
64-
The extensibility project will document the most widely adopted extensions and those best suited for particular areas as official **Extensions to MCP** to encourage broader ecosystem adoption.
58+
Establishing officially endorsed extensions to MCP. Patterns are emerging in the MCP ecosystem that are highly relevant to specific use cases or industry domains. The project will document the most widely adopted extensions and those best suited for particular areas as official _Extensions to MCP_ to encourage broader ecosystem adoption.
6559

6660
### SDK Support Standardization
6761

68-
We will be introducing a tiering system for MCP SDKs to provide clarity about support levels and spec conformance. This standardization will be based on factors including specification compliance velocity, maintainer responsiveness, and feature completeness. The tiering framework will help users make informed decisions when selecting SDKs for their projects while also having clear expectations for different levels of support they can get from taking on the dependency.
62+
Introducing a tiering system for MCP SDKs to provide clarity to developers about support levels and maintenance quality. This standardization will be based on factors including specification compliance speed, maintenance responsiveness, and feature completeness. The tiering framework will help users make informed decisions when selecting SDKs for their projects while establishing clear expectations for different levels of support.
6963

7064
## Call for Contributors
7165

72-
As MCP grows, we welcome contributors - both individuals and companies. Contributions are especially needed in several areas key to the MCP success:
66+
The project welcomes contributors, both individuals and companies. Contributions are particularly needed in several key areas:
7367

7468
### SDK Maintenance
7569

76-
- [**TypeScript SDK**](https://github.com/modelcontextprotocol/typescript-sdk) - Needs additional maintainers for feature development and bug fixes
77-
- [**Swift SDK**](https://github.com/modelcontextprotocol/swift-sdk) - Requires attention for Apple ecosystem support
78-
- [Other language SDKs](https://modelcontextprotocol.io/docs/sdk) welcome continued contributions
70+
- [TypeScript SDK](https://github.com/modelcontextprotocol/typescript-sdk) - Needs additional maintainers for feature development and bug fixes
71+
- [Swift SDK](https://github.com/modelcontextprotocol/swift-sdk) - Requires attention for Apple ecosystem support
72+
- [Other language SDKs](https://modelcontextprotocol.io/docs/sdk) - Continued contributions are welcome
7973

8074
### Tooling
8175

82-
- [**Inspector**](https://github.com/modelcontextprotocol/inspector) - Development and maintenance of debugging tools
83-
- [**Registry**](https://github.com/modelcontextprotocol/registry) - Backend API and CLI development for the server registry; in particular, Go expertise would be very welcome
76+
- [Inspector](https://github.com/modelcontextprotocol/inspector) - Maintenance of the MCP Inspector
77+
- [Registry](https://github.com/modelcontextprotocol/registry/tree/main/docs) - Backend API and CLI development for the server registry; in particular, Go expertise would be very welcome. Front-end TypeScript development expertise may also be helpful in the near future.
8478

8579
## Input from Client Implementors
8680

87-
MCP is also not just about servers, but also clients - the tools that enable customers to connect to our ecosystem. If you are implementing MCP clients, your experience and understanding of requirements are crucial for shaping the protocol's evolution. We want to hear from you on both protocol capabilities as well as processes that will enable you to scale your work as we add new features.
81+
We particularly value feedback from teams implementing widely used MCP clients. Your experience and requirements are crucial for shaping the protocol's evolution.
8882

8983
Client implementors are invited to join the `#client-implementors` working group channel in the [MCP Discord](https://modelcontextprotocol.io/community/communication).
84+
85+
Thank you for your continued support and we look forward to sharing more soon.

0 commit comments

Comments
 (0)