Merge branch 'main' into feat/task-jumpscares

Author: LucaButBoring

.gitattributes

@@ -2,3 +2,5 @@ package-lock.json linguist-generated=true
 schema/*/schema.json linguist-generated=true
 docs/specification/*/schema.md linguist-generated=true
 docs/specification/*/schema.mdx linguist-generated=true
+docs/community/seps/index.mdx linguist-generated=true
+docs/community/seps/[0-9]*.mdx linguist-generated=true

.github/dependabot.yml

@@ -4,3 +4,8 @@ updates:
     directory: '/'
     schedule:
       interval: 'weekly'
+
+  - package-ecosystem: 'npm'
+    directory: '/'
+    schedule:
+      interval: 'weekly'

.github/workflows/sep-reminder.yml

@@ -11,7 +11,7 @@ permissions: {}
 jobs:
   discord-reminder:
     runs-on: ubuntu-latest
-
+    if: github.repository == 'modelcontextprotocol/modelcontextprotocol'
     steps:
       - name: Post Discord reminder
         run: |

.gitignore

@@ -2,3 +2,4 @@
 node_modules/
 .DS_Store
 .idea/
+.jj

AGENTS.md

@@ -9,6 +9,13 @@ This repository contains the MCP specification, documentation, and blog.
   - `docs/specification/` - MCP specification (more formal, versioned)
 - `blog/` - Hugo blog (`npm run serve:blog`)
 
+### Documentation Guidelines
+
+- When creating flowcharts, and graphs to visualize aspect of the protocol, use mermaid diagrams where
+  possible.
+- When writing tables, ensure column headers and columns are aligned with whitespace.
+- Before pushing or creating PR's ensure that `npm run prep` is free of warnings and errors.
+
 ## Specification Versioning
 
 Specifications use **date-based versioning** (YYYY-MM-DD), not semantic versioning:

MAINTAINERS.md

@@ -111,13 +111,15 @@ This document lists current maintainers in the Model Context Protocol project.
 - [Cliff Hall](https://github.com/cliffhall)
 - [Konstantin Konstantinov](https://github.com/KKonstantinov)
 - [Ola Hungerford](https://github.com/olaservo)
+- [Bob Dickinson](https://github.com/BobDickinson)
 
 ### Registry
 
 - [Toby Padilla](https://github.com/toby)
 - [Tadas Antanavicius](https://github.com/tadasant)
 - [Adam Jones](https://github.com/domdomegg)
 - [Radoslav (Rado) Dimitrov](https://github.com/rdimitrov)
+- [Bob Dickinson](https://github.com/BobDickinson)
 
 ### MCPB (Model Context Protocol Bundle)
 

SECURITY.md

@@ -1,24 +1,24 @@
 # Security Policy
 
-Thank you for helping us keep the SDKs and systems they interact with secure.
+Thank you for helping keep the Model Context Protocol and its ecosystem secure.
 
 ## Reporting Security Issues
 
-This SDK is maintained by [Anthropic](https://www.anthropic.com/) as part of the Model
-Context Protocol project.
+If you discover a security vulnerability in this repository, please report it through
+the [GitHub Security Advisory process](https://docs.github.com/en/code-security/security-advisories/guidance-on-reporting-and-writing-information-about-vulnerabilities/privately-reporting-a-security-vulnerability)
+for this repository.
 
-The security of our systems and user data is Anthropic's top priority. We appreciate the
-work of security researchers acting in good faith in identifying and reporting potential
-vulnerabilities.
+Please **do not** report security vulnerabilities through public GitHub issues, discussions,
+or pull requests.
 
-Our security program is managed on HackerOne and we ask that any validated vulnerability
-in this functionality be reported through their
-[submission form](https://hackerone.com/anthropic-vdp/reports/new?type=team&report_type=vulnerability).
+## What to Include
 
-## Vulnerability Disclosure Program
+To help us triage and respond quickly, please include:
 
-Our Vulnerability Program Guidelines are defined on our
-[HackerOne program page](https://hackerone.com/anthropic-vdp).
+- A description of the vulnerability
+- Steps to reproduce the issue
+- The potential impact
+- Any suggested fixes (optional)
 
 ## Intended Behaviors and Trust Model
 
@@ -57,7 +57,7 @@ to certain servers based on configuration; users should review these settings.
 ### Behaviors That Are Not Vulnerabilities
 
 The following behaviors are intentional features of MCP and are **not** eligible for
-security vulnerability reports or bug bounty rewards:
+security vulnerability reports:
 
 #### Command Execution for STDIO Transport
 

docs/clients.mdx

@@ -5,7 +5,7 @@ description: "A list of applications that support MCP integrations"
 
 {/* prettier-ignore-start */}
 
-export const FEATURES = ["Resources", "Prompts", "Tools", "Discovery", "Instructions", "Sampling", "Roots", "Elicitation", "Tasks", "Apps"];
+export const FEATURES = ["Resources", "Prompts", "Tools", "Discovery", "Instructions", "Sampling", "Roots", "Elicitation", "CIMD", "DCR", "Tasks", "Apps"];
 
 export const FEATURE_COLORS = {
   Resources: "blue",
@@ -18,6 +18,8 @@ export const FEATURE_COLORS = {
   Elicitation: "green",
   Tasks: "orange",
   Apps: "orange",
+  DCR: "yellow",
+  CIMD: "yellow",
 };
 
 export const FeatureBadge = ({ feature }) => {
@@ -279,18 +281,20 @@ export const McpClient = ({ name, homepage, supports, sourceCode, instructions,
 
 This page showcases applications that support the Model Context Protocol (MCP). Each client may support different MCP features:
 
-| Feature                                 | Description                                               |
-| --------------------------------------- | --------------------------------------------------------- |
-| <FeatureBadge feature="Resources" />    | Server-exposed data and content                           |
-| <FeatureBadge feature="Prompts" />      | Pre-defined templates for LLM interactions                |
-| <FeatureBadge feature="Tools" />        | Executable functions that LLMs can invoke                 |
-| <FeatureBadge feature="Discovery" />    | Support for tools/prompts/resources changed notifications |
-| <FeatureBadge feature="Instructions" /> | Server-provided guidance for LLMs                         |
-| <FeatureBadge feature="Sampling" />     | Server-initiated LLM completions                          |
-| <FeatureBadge feature="Roots" />        | Filesystem boundary definitions                           |
-| <FeatureBadge feature="Elicitation" />  | User information requests                                 |
-| <FeatureBadge feature="Tasks" />        | Long-running operation tracking                           |
-| <FeatureBadge feature="Apps" />         | Interactive HTML interfaces                               |
+| Feature                                 | Description                                                                                                  |
+| --------------------------------------- | ------------------------------------------------------------------------------------------------------------ |
+| <FeatureBadge feature="Resources" />    | Server-exposed data and content                                                                              |
+| <FeatureBadge feature="Prompts" />      | Pre-defined templates for LLM interactions                                                                   |
+| <FeatureBadge feature="Tools" />        | Executable functions that LLMs can invoke                                                                    |
+| <FeatureBadge feature="Discovery" />    | Support for tools/prompts/resources changed notifications                                                    |
+| <FeatureBadge feature="Instructions" /> | Server-provided guidance for LLMs                                                                            |
+| <FeatureBadge feature="Sampling" />     | Server-initiated LLM completions                                                                             |
+| <FeatureBadge feature="Roots" />        | Filesystem boundary definitions                                                                              |
+| <FeatureBadge feature="Elicitation" />  | User information requests                                                                                    |
+| <FeatureBadge feature="CIMD" />         | [Client ID Metadata Document](specification/latest/basic/authorization#client-id-metadata-documents) support |
+| <FeatureBadge feature="DCR" />          | [Dynamic Client Registration](specification/latest/basic/authorization#dynamic-client-registration) support  |
+| <FeatureBadge feature="Tasks" />        | Long-running operation tracking                                                                              |
+| <FeatureBadge feature="Apps" />         | Interactive HTML interfaces                                                                                  |
 
 <Note>
 
@@ -631,7 +635,7 @@ ChatFrame is a cross-platform desktop chatbot that unifies access to multiple AI
 <McpClient
   name="ChatGPT"
   homepage="https://chatgpt.com"
-  supports="Tools"
+  supports="Tools, DCR"
   instructions="https://platform.openai.com/docs/guides/developer-mode"
 >
 
@@ -681,7 +685,7 @@ Chorus is a native Mac app for chatting with AIs. Chat with multiple models at o
 <McpClient
   name="Claude Code"
   homepage="https://claude.com/product/claude-code"
-  supports="Resources, Prompts, Tools, Roots, Instructions, Discovery"
+  supports="Resources, Prompts, Tools, Roots, Instructions, Discovery, DCR"
   instructions="https://code.claude.com/docs/en/mcp"
 >
 
@@ -697,7 +701,7 @@ Claude Code is an interactive agentic coding tool from Anthropic that helps you
 <McpClient
   name="Claude Desktop App"
   homepage="https://claude.ai/download"
-  supports="Resources, Prompts, Tools, Apps"
+  supports="Resources, Prompts, Tools, Apps, DCR"
   instructions={[
     ["Local servers", "https://support.claude.com/en/articles/10949351-getting-started-with-local-mcp-servers-on-claude-desktop"],
     ["Remote servers", "https://support.claude.com/en/articles/11175166-getting-started-with-custom-connectors-using-remote-mcp"]
@@ -718,7 +722,7 @@ Claude Desktop provides comprehensive support for MCP, enabling deep integration
 <McpClient
   name="Claude.ai"
   homepage="https://claude.ai"
-  supports="Resources, Prompts, Tools"
+  supports="Resources, Prompts, Tools, CIMD, DCR"
 >
 
 Claude.ai is Anthropic's web-based AI assistant that provides MCP support for remote servers.
@@ -825,7 +829,7 @@ Copilot-MCP enables AI coding assistance via MCP.
 <McpClient
   name="Cursor"
   homepage="https://docs.cursor.com/context/mcp#protocol-support"
-  supports="Prompts, Tools, Roots, Elicitation"
+  supports="Prompts, Tools, Roots, Elicitation, DCR"
   instructions="https://docs.cursor.com/context/model-context-protocol"
 >
 
@@ -970,7 +974,7 @@ Think n8n + ChatGPT. FLUJO is a desktop application that integrates with MCP to
 <McpClient
   name="Gemini CLI"
   homepage="https://github.com/google-gemini/gemini-cli"
-  supports="Prompts, Tools, Instructions"
+  supports="Prompts, Tools, Instructions, DCR"
   instructions="https://geminicli.com/docs/tools/mcp-server/"
 >
 
@@ -1015,7 +1019,7 @@ Genkit is a cross-language SDK for building and integrating GenAI features into
 <McpClient
   name="GitHub Copilot coding agent"
   homepage="https://docs.github.com/en/copilot/concepts/about-copilot-coding-agent"
-  supports="Tools"
+  supports="Tools, DCR"
 >
 
 Delegate tasks to GitHub Copilot coding agent and let it work in the background while you stay focused on the highest-impact and most interesting work
@@ -1101,6 +1105,24 @@ HyperAgent is Playwright supercharged with AI. With HyperAgent, you no longer ne
 
 </McpClient>
 
+<McpClient
+  name="Inspector"
+  homepage="https://tryinspector.com"
+  supports="Tools, Prompts, Resources, DCR"
+  instructions="https://tryinspector.com/docs"
+>
+
+Inspector is a visual editor for your codebase. It connects to Cursor, Claude Code, and Codex so you can edit your frontend visually. Move elements, change text, and ship real code without touching CSS.
+
+**Key features:**
+
+- Design Mode: Move elements, edit text, and zoom in to interact with your front-end like Figma.
+- Agent Connect: Plug in Cursor, Claude Code, or Codex.
+- Version Control: Stage changes and open PRs from Inspector.
+- MCP Client: Connect any MCP Server you want!
+
+</McpClient>
+
 <McpClient
   name="Jenova"
   homepage="https://jenova.ai"
@@ -1238,7 +1260,7 @@ Langflow is an open-source visual builder that lets developers rapidly prototype
 <McpClient
   name="LibreChat"
   homepage="https://github.com/danny-avila/LibreChat"
-  supports="Tools, Instructions"
+  supports="Tools, Instructions, DCR"
   instructions="https://www.librechat.ai/docs/features/mcp"
 >
 
@@ -1451,7 +1473,7 @@ MCPHub is a powerful Neovim plugin that integrates MCP (Model Context Protocol)
 <McpClient
   name="MCPJam"
   homepage="https://github.com/MCPJam/inspector"
-  supports="Resources, Prompts, Tools, Elicitation, Instructions, Tasks, Apps"
+  supports="Resources, Prompts, Tools, Elicitation, Instructions, Tasks, Apps, CIMD, DCR"
   instructions="https://docs.mcpjam.com/getting-started"
 >
 
@@ -1511,7 +1533,7 @@ Memex is the first MCP client and MCP server builder - all-in-one desktop app. U
 <McpClient
   name="Memgraph Lab"
   homepage="https://memgraph.com/lab"
-  supports="Tools, Sampling, Elicitation, Instructions"
+  supports="Resources, Prompts, Tools, Sampling, Elicitation, Instructions"
   instructions="https://memgraph.com/docs/memgraph-lab/features/graphchat#mcp-servers"
 >
 
@@ -1521,7 +1543,7 @@ Memex is the first MCP client and MCP server builder - all-in-one desktop app. U
 
 - Build GraphRAG workflows powered by knowledge graphs as the data backbone
 - Connect remote MCP servers via `SSE` or `Streamable HTTP`
-- Support for MCP tools, sampling, elicitation, and instructions
+- Support for MCP resources, prompts, tools, sampling, elicitation, and instructions
 - Create multiple agents with different configurations for easy comparison and debugging
 - Works with various LLM providers (OpenAI, Azure OpenAI, Anthropic, Gemini, Ollama, DeepSeek)
 - Available as a Desktop app or Docker container
@@ -1803,7 +1825,7 @@ RecurseChat is a powerful, fast, local-first chat client with MCP support. Recur
 <McpClient
   name="Replit"
   homepage="https://replit.com/products/agent"
-  supports="Tools"
+  supports="Tools, DCR"
 >
 
 Replit Agent is an AI-powered software development tool that builds and deploys applications through natural language. It supports MCP integration, enabling users to extend the agent's capabilities with custom tools and data sources.
@@ -2135,7 +2157,7 @@ v0 turns your ideas into fullstack apps, no code required. Describe what you wan
 <McpClient
   name="VS Code GitHub Copilot"
   homepage="https://code.visualstudio.com/"
-  supports="Resources, Prompts, Tools, Discovery, Sampling, Roots, Elicitation, Instructions, Apps"
+  supports="Resources, Prompts, Tools, Discovery, Sampling, Roots, Elicitation, Instructions, Apps, CIMD, DCR"
   instructions="https://code.visualstudio.com/docs/copilot/customization/mcp-servers"
 >