Slack MCP Server
A Model Context Protocol (MCP) server for Slack integration, allowing Claude to interact with your Slack workspace.
README
Slack MCP Server
A Model Context Protocol (MCP) server for Slack integration, allowing Claude to interact with your Slack workspace. This server runs exclusively in Docker/Podman containers.
Features
- Send messages to channels or users
- Send replies to message threads
- Search for messages across the workspace
- List all channels in workspace
- Get channel message history
- Get replies from message threads
- List all users in workspace
Prerequisites
- Docker or Podman installed
- A Slack Bot Token (see Setup section)
Setup
1. Create a Slack App
- Go to https://api.slack.com/apps
- Click "Create New App" → "From scratch"
- Give it a name and select your workspace
2. Configure Bot Token Scopes
Go to OAuth & Permissions and add these Bot Token Scopes:
channels:readchat:writeusers:readchannels:historygroups:readgroups:history
3. Install the App
- Click "Install to Workspace"
- Copy the Bot User OAuth Token (starts with
xoxb-)
4. Set Up Environment Variable
Create a .env file in the project root:
SLACK_BOT_TOKEN=xoxb-your-bot-token-here
Running with Docker
Build the Image
docker build -t slack-mcp-server .
Run the Container
docker run -i \
-e SLACK_BOT_TOKEN=xoxb-your-bot-token-here \
slack-mcp-server
Using docker-compose
# Set your token in .env file first
docker-compose up
Running with Podman
Build the Image
podman build -t slack-mcp-server .
Run the Container
podman run -i \
-e SLACK_BOT_TOKEN=xoxb-your-bot-token-here \
slack-mcp-server
Using podman-compose
# Set your token in .env file first
podman-compose up
Configuration with Claude Code
Add this to your Claude Code MCP settings file:
Docker Configuration
{
"mcpServers": {
"slack": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"SLACK_BOT_TOKEN=xoxb-your-actual-bot-token-from-step-3",
"slack-mcp-server"
]
}
}
}
Podman Configuration
{
"mcpServers": {
"slack": {
"command": "podman",
"args": [
"run",
"-i",
"--rm",
"-e",
"SLACK_BOT_TOKEN=xoxb-your-actual-bot-token-from-step-3",
"slack-mcp-server"
]
}
}
}
Alternative: Using Environment File
{
"mcpServers": {
"slack": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"--env-file",
"/absolute/path/to/.env",
"slack-mcp-server"
]
}
}
}
Available Tools
The server communicates via stdio and provides these tools:
send_message: Send a message to a channel or usersend_thread_reply: Send a reply to a message threadsearch_messages: Search for messages across the workspace (supports Slack search syntax)list_channels: List all workspace channelsget_channel_history: Get recent messages from a channelget_thread_replies: Get all replies from a message threadlist_users: List all workspace users
Search Examples
The search_messages tool supports Slack's powerful search syntax:
"error message"- Search for exact phrasefrom:@username- Messages from specific userin:#channel- Messages in specific channelafter:2024-01-01- Messages after datehas:link- Messages containing links- Combine:
from:@john in:#general error- Complex queries
Development
Build and Test Locally
# Build the image
docker build -t slack-mcp-server:dev .
# Run tests
docker run --rm slack-mcp-server:dev npm test
Shell Access for Debugging
docker run -it --rm \
-e SLACK_BOT_TOKEN=xoxb-your-token \
--entrypoint /bin/sh \
slack-mcp-server
Security Features
- Multi-stage build for minimal image size
- Non-root user execution
- Read-only root filesystem
- No new privileges
- Resource limits configured in docker-compose
- Minimal Alpine Linux base image
Troubleshooting
Container exits immediately
Make sure you're running with -i (interactive) flag for stdin communication.
Token not found
Verify your SLACK_BOT_TOKEN is correctly set in the environment or .env file.
Permission errors
The container runs as a non-root user (nodejs). Ensure any mounted volumes have appropriate permissions.
License
MIT
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。