Other

MCP Celebrates First Anniversary with Major November 2025 Specification Update

Model Context Protocol (MCP), developed by Anthropic, celebrates its first anniversary with a major specification update. With approximately 2,000 server entries and 407% growth since September, major companies including GitHub, OpenAI, Microsoft, AWS, and Google Cloud have expressed their support.

MCP Anthropic Open Source Agent AI Developer Tools Ecosystem

Model Context Protocol (MCP), developed by Anthropic, has celebrated its first anniversary in November 2025 with a major specification update release.

Remarkable Growth

In just one year since its November 2024 announcement, MCP has evolved from a small open-source experiment into an industry standard.

Key Statistics

MetricValue
MCP Registry Server Count~2,000
Growth Since September407%
Maintainers58 (9 core leaders)
Discord Contributors2,900+
Weekly New Contributors100+

The ecosystem has expanded to the point where “if you can think of a scenario, there’s probably an MCP server for it.”

Major Companies Building MCP Servers

Leading tech companies are actively building MCP servers:

  • Notion: Note management
  • Stripe: Payment workflows
  • GitHub: Engineering process automation
  • Hugging Face: Model management
  • Postman: API test automation

November 2025 New Specifications

Task-Based Workflows (SEP-1686)

A new feature enabling server work tracking. Supports the following state management:

  • working: Processing
  • input_required: Awaiting input
  • completed: Completed
  • failed: Failed
  • cancelled: Cancelled

Designed for healthcare data analysis and multi-step enterprise workflows.

Simplified Authorization Flow (SEP-991)

URL-based client registration significantly reduces the complexity of Dynamic Client Registration (DCR). Enables more streamlined authentication using OAuth client ID metadata documents.

Security Enhancements

  • SEP-1024: Client security requirements for local server installation
  • SEP-835: Default scope definitions for authorization specification
  • Enterprise MCP registry vision established

Extensions

Optional, configurable extensions support experimental features while keeping the core protocol stable:

  • SEP-1046: OAuth client authentication
  • SEP-990: Cross App Access

URL Mode Invocation (SEP-1036)

Securely collects authentication credentials through browser OAuth flows. Improves security by ensuring clients never see authentication information directly.

Sampling with Tools (SEP-1577)

A groundbreaking feature enabling servers to run their own agent loops:

  • Tool invocation
  • Parallel execution
  • Multi-step reasoning

Enables more complex agent workflows.

Developer Experience Improvements

  • SEP-986: Standard format for tool names
  • SEP-1319: Decoupling from RPC methods
  • SEP-1699: SSE polling with server-side disconnection
  • SEP-1309: Improved SDK specification versioning

Industry Partner Support

Major companies have expressed support for MCP’s standardization and importance in agent development:

  • GitHub
  • OpenAI
  • Block
  • Microsoft
  • Hugging Face
  • Okta
  • AWS
  • Google Cloud
  • Obot AI

Governance Structure Established

A system where community and Anthropic maintainers collaborate to approve Specification Extension Proposals (SEPs) has been established. Formal working groups and interest groups have been introduced, with a clear policy that “the maintainer team is not a gatekeeper, but helps surface issues and align on solutions.”

Future Direction

This release maintains backward compatibility. The development team plans to focus on:

  • Improved Reliability & Observability: More stable protocol operation
  • Better Server Configuration Patterns: Enhanced developer experience
  • Refined Enterprise Security Models: Accelerated enterprise adoption

Summary

MCP’s first anniversary marks an important milestone in AI agent interoperability:

  • Rapid Growth: 2,000 servers and 407% growth rate in one year
  • Industry Standardization: Support from major tech companies
  • Technical Maturity: Major updates including task-based workflows and security enhancements
  • Open Governance: Community-driven specification development

As AI agents become mainstream, MCP is establishing itself as the foundational infrastructure for the ecosystem.