mcp-server-matrix

mcp-server-matrix

Enables AI assistants to interact with Matrix, the open decentralized communication protocol, allowing them to send and read messages, manage rooms, and perform other Matrix operations through natural language.

Category
访问服务器

README

mcp-server-matrix

README на русском

A Model Context Protocol (MCP) server for Matrix — the open, decentralized communication protocol.

Built on matrix-nio, this server lets any MCP-compatible AI assistant (Claude, OpenClaw, etc.) send and read messages, manage rooms, and interact with any Matrix homeserver (Synapse, Dendrite, Conduit).

Features

Tool Description
send_message Send a message (text, HTML, or Markdown) with optional reply
read_messages Read recent messages (with pagination)
get_messages_by_date Get messages filtered by date range
list_rooms List all joined rooms
get_room_info Room details — name, topic, members, encryption status
get_room_members List joined members with display names and avatars
get_notification_counts Unread and mention counts per room
get_direct_messages List DM conversations with unread status
send_direct_message Send a DM (auto-creates room if needed)
get_user_profile User profile — display name, avatar
join_room Join a room by ID or alias
leave_room Leave a room
create_room Create a new room (with optional invites)
invite_user Invite a user to a room
send_reaction React to a message with an emoji
resolve_alias Resolve #alias:server to a room ID

Installation

pip install mcp-server-matrix

Or with uv:

uvx mcp-server-matrix

Configuration

Set environment variables:

Variable Required Description
MATRIX_HOMESERVER Yes Homeserver URL (e.g. https://matrix.org)
MATRIX_USER Yes Full user ID (e.g. @bot:matrix.org)
MATRIX_PASSWORD Yes* Account password
MATRIX_ACCESS_TOKEN Yes* Access token (alternative to password)
MATRIX_STORE_PATH No Path for nio session store (default: ~/.mcp-server-matrix/nio_store/)
MCP_LOG_LEVEL No Log level: DEBUG, INFO, WARNING, ERROR (default: INFO)

* Provide either MATRIX_PASSWORD or MATRIX_ACCESS_TOKEN.

Usage

Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "matrix": {
      "command": "mcp-server-matrix",
      "env": {
        "MATRIX_HOMESERVER": "https://matrix.example.com",
        "MATRIX_USER": "@bot:example.com",
        "MATRIX_PASSWORD": "your-password"
      }
    }
  }
}

Claude Code

{
  "mcpServers": {
    "matrix": {
      "command": "mcp-server-matrix",
      "env": {
        "MATRIX_HOMESERVER": "https://matrix.example.com",
        "MATRIX_USER": "@bot:example.com",
        "MATRIX_ACCESS_TOKEN": "syt_..."
      }
    }
  }
}

OpenClaw

Add to openclaw.json:

{
  "mcp": {
    "servers": {
      "matrix": {
        "command": "mcp-server-matrix",
        "env": {
          "MATRIX_HOMESERVER": "https://matrix.example.com",
          "MATRIX_USER": "@bot:example.com",
          "MATRIX_PASSWORD": "your-password"
        }
      }
    }
  }
}

Direct (stdio)

export MATRIX_HOMESERVER=https://matrix.example.com
export MATRIX_USER=@bot:example.com
export MATRIX_PASSWORD=your-password

mcp-server-matrix

Examples

Once connected, your AI assistant can:

  • Read messages: "What are the latest messages in #general?"
  • Send messages: "Send 'Hello everyone!' to #announcements"
  • Manage rooms: "Create a room called 'Project X' and invite @alice:matrix.org"
  • Get info: "How many members are in #team-chat?"

Supported Homeservers

Tested with:

  • Synapse (reference implementation)
  • Should work with any spec-compliant homeserver (Dendrite, Conduit, etc.)

Requirements

  • Python 3.10+
  • A Matrix account on any homeserver
  • Network access to the homeserver

License

MIT

Links

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选