Discord MCP Server
Enables AI assistants to manage a Discord server, including creating channels, categories, roles, setting permissions, and assigning roles through natural language commands.
README
Discord MCP Server
Manage a Discord server directly from your AI coding assistant. Create channels, categories, roles, set permissions, assign roles, and more — all through natural language.
Tools
| Tool | Description |
|---|---|
create_category |
Create a new category channel |
create_channel |
Create a text or voice channel under a category |
create_role |
Create a new role with color, hoist, mentionable |
edit_role |
Edit a role's name, color, hoist, or mentionable |
delete_role |
Delete a role (non-managed, below bot's hierarchy) |
assign_role |
Assign a role to a member by username |
remove_role |
Remove a role from a member by username |
restrict_channel |
Set channel permissions (allow/deny roles, read-only mode) |
create_role_restricted_channel |
Create a channel + restrict it in one call |
delete_category |
Delete a category channel |
delete_channel |
Delete a text or voice channel |
Setup
1. Create a Discord bot
- Go to https://discord.com/developers/applications
- Click New Application → name it → Create
- Go to Bot → Add Bot
- Under Privileged Gateway Intents, enable:
Server Members IntentMessage Content Intent
- Copy the Token
- Go to OAuth2 → URL Generator
- Scopes:
bot - Bot Permissions:
Administrator(or manually select: Manage Channels, Manage Roles, Kick Members, Manage Webhooks, Send Messages, Read Messages, View Channels)
- Scopes:
- Open the generated URL in your browser to invite the bot to your server
2. Configure environment
cp .env.example .env
Edit .env and paste your bot token:
DISCORD_TOKEN=your_token_here
3. Install dependencies
pip install -r requirements.txt
MCP Host Configuration
opencode
Add to opencode.json in your project root:
{
"mcp": {
"discord": {
"type": "local",
"command": ["python", "/absolute/path/to/discord_mcp_server.py"]
}
}
}
Restart opencode to load the server.
Claude Desktop
Edit claude_desktop_config.json (File → Settings → Developer → Edit Config):
{
"mcpServers": {
"discord": {
"command": "python",
"args": ["/absolute/path/to/discord_mcp_server.py"],
"env": {
"DISCORD_TOKEN": "your_token_here"
}
}
}
}
Claude Code
Add to ~/.claude.json or .mcp.json in your project:
{
"mcpServers": {
"discord": {
"command": "python",
"args": ["/absolute/path/to/discord_mcp_server.py"],
"env": {
"DISCORD_TOKEN": "your_token_here"
}
}
}
}
Cursor
Create .cursor/mcp.json in your project root (or ~/.cursor/mcp.json for global):
{
"mcpServers": {
"discord": {
"command": "python",
"args": ["/absolute/path/to/discord_mcp_server.py"],
"env": {
"DISCORD_TOKEN": "your_token_here"
}
}
}
}
Windsurf
Add to ~/.codeium/windsurf/mcp_config.json:
{
"mcpServers": {
"discord": {
"command": "python",
"args": ["/absolute/path/to/discord_mcp_server.py"],
"env": {
"DISCORD_TOKEN": "your_token_here"
}
}
}
}
Codex CLI
Add to ~/.codex/config.toml:
[mcp_servers.discord]
command = "python"
args = ["/absolute/path/to/discord_mcp_server.py"]
[mcp_servers.discord.env]
DISCORD_TOKEN = "your_token_here"
Or add via CLI:
codex mcp add discord -- python /absolute/path/to/discord_mcp_server.py
Antigravity (Google)
Add to ~/.gemini/config/mcp_config.json:
{
"mcpServers": {
"discord": {
"command": "python",
"args": ["C:\\path\\to\\discord_mcp_server.py"],
"env": {
"DISCORD_TOKEN": "your_token_here"
}
}
}
}
Open Antigravity → Agent panel → ... → MCP Servers → Manage MCP Servers → View raw config.
Note: If using the
.envfile instead of inlineenv, omit theenvblock and ensureload_dotenv()in the script finds your.env(the working directory depends on the host).
How it works
The script runs an MCP server over stdio that connects to Discord via the bot token. It registers all tools on startup, and your AI assistant calls them as needed. The Discord bot only needs to be in one server — it auto-selects the first guild on login.
推荐服务器
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。
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。