Slack MCP Server
Enables comprehensive Slack workspace integration through AI assistants, allowing users to manage channels, send messages, upload files, search conversations, and interact with users through natural language commands.
README
Slack MCP Server
A TypeScript-based Model Context Protocol (MCP) server that provides comprehensive Slack workspace integration. This server enables AI assistants to interact with Slack channels, users, messages, files, and more through the MCP protocol.
Table of Contents
Features
Channel Management
- list_channels - List workspace channels with filtering options
- get_channel_info - Get detailed information about a specific channel
- create_channel - Create new public or private channels
- archive_channel - Archive channels
User Management
- list_users - List all users in the workspace
- get_user_info - Get detailed user profile information
- invite_to_channel - Invite users to channels
Messaging
- send_message - Send text messages to channels
- update_message - Update existing messages
- delete_message - Delete messages
- get_channel_history - Fetch message history with pagination
- search_messages - Search messages across the workspace
- send_formatted_message - Send messages with Block Kit formatting
File Operations
- upload_file - Upload files to channels
Reactions
- add_reaction - Add emoji reactions to messages
- remove_reaction - Remove emoji reactions
Workspace Information
- get_team_info - Get workspace/team information
Prerequisites
Before you begin, ensure you have:
- Node.js 18 or higher - Verify with:
node --version - npm - Verify with:
npm --version - Slack workspace admin access - You need permission to create and install apps
- Estimated setup time: ~20 minutes
Installation & Setup
Step 1: Project Setup (5 minutes)
Clone the repository and install dependencies:
# Clone the repository
git clone <your-repo-url>
cd slack-bot-mcp
# Install dependencies
npm install
✓ Verification: Run npm list @modelcontextprotocol/sdk to confirm the MCP SDK is installed.
Step 2: Create Slack App (10 minutes)
2.1 Create the App
- Go to https://api.slack.com/apps
- Click "Create New App" → "From scratch"
- Enter an app name (e.g., "MCP Bot") and select your workspace
- Click "Create App"
2.2 Configure OAuth Scopes
Navigate to "OAuth & Permissions" in the sidebar, scroll to "Scopes", and add these Bot Token Scopes:
| Scope | Purpose | Used By |
|---|---|---|
channels:read |
View basic channel information | list_channels, get_channel_info |
channels:write |
Create channels | create_channel |
channels:manage |
Archive channels, invite users | archive_channel, invite_to_channel |
users:read |
View user profiles | list_users, get_user_info |
chat:write |
Send and update messages | send_message, update_message, delete_message |
files:write |
Upload files to Slack | upload_file |
reactions:write |
Add/remove emoji reactions | add_reaction, remove_reaction |
team:read |
Get workspace information | get_team_info |
search:read |
Search messages | search_messages |
💡 Tip: Each scope is necessary for specific tools to function. Missing scopes will cause "missing_scope" errors.
2.3 Install to Workspace
- Scroll to "OAuth Tokens for Your Workspace"
- Click "Install to Workspace"
- Review permissions and click "Allow"
2.4 Copy Your Bot Token
- After installation, you'll see the "Bot User OAuth Token"
- Click "Copy" (token starts with
xoxb-) - Keep this token secure - you'll need it in the next step
✓ Verification: Your token should look like: xoxb-<numbers>-<numbers>-<random-string>
Step 3: Configure Environment (2 minutes)
Create your environment configuration:
# Copy the example file
cp .env.example .env
Edit .env and add your bot token:
SLACK_BOT_TOKEN=xoxb-your-actual-token-here
✓ Verification: Run this command to confirm your token is loaded:
node -e "require('dotenv').config(); console.log('Token loaded:', !!process.env.SLACK_BOT_TOKEN)"
You should see: Token loaded: true
⚠️ Common Issue: Make sure your token starts with
xoxb-(notxoxp-orxoxa-)
Step 4: Build & Verify (3 minutes)
Build the TypeScript project:
npm run build
✓ Verification: Confirm the build/ directory was created:
ls build/index.js
You should see the compiled index.js file.
Testing Your Setup
Before integrating with Claude Desktop, test that everything works:
Using MCP Inspector
The MCP Inspector provides a GUI to test your tools:
npx @modelcontextprotocol/inspector node build/index.js
This will open a web interface where you can:
- See all available tools
- Test individual tools with sample inputs
- View responses and debug errors
Quick Test Commands
Try these simple tests in the MCP Inspector:
-
List channels:
- Tool:
list_channels - Arguments:
{"limit": 5} - Expected: List of your workspace channels
- Tool:
-
Get team info:
- Tool:
get_team_info - Arguments:
{} - Expected: Your workspace name and details
- Tool:
-
List users:
- Tool:
list_users - Arguments:
{"limit": 5} - Expected: List of workspace users
- Tool:
✓ Success Indicator: If these commands return data (not errors), your setup is working correctly!
⚠️ Troubleshooting: If you see "invalid_auth" or "token_revoked", double-check your
SLACK_BOT_TOKENin the.envfile.
Usage
With Claude Desktop
Add to your Claude Desktop configuration file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%/Claude/claude_desktop_config.json
Option 1: Direct from GitHub (Recommended)
Use npx to run directly from the GitHub repository:
{
"mcpServers": {
"slack": {
"command": "npx",
"args": [
"github:Hais/slack-bot-mcp"
],
"env": {
"SLACK_BOT_TOKEN": "xoxb-your-token-here"
}
}
}
}
Benefits:
- ✓ No local installation needed
- ✓ Always uses the latest version
- ✓ Simpler configuration
- ✓ Works on any machine with Node.js
Option 2: Local Installation
For local development or offline use:
{
"mcpServers": {
"slack": {
"command": "node",
"args": ["/absolute/path/to/slack-bot-mcp/build/index.js"],
"env": {
"SLACK_BOT_TOKEN": "xoxb-your-token-here"
}
}
}
}
Setup for local installation:
- Clone and build the project (see Installation & Setup)
- Replace
/absolute/path/to/slack-bot-mcpwith your full project path- Find it by running
pwdin your project directory - Example:
/Users/yourname/projects/slack-bot-mcp/build/index.js
- Find it by running
Important notes:
-
Token configuration: Set
SLACK_BOT_TOKENin the config file as shown above- The config file token takes precedence over
.envfiles - Keep your token secure and never commit it to version control
- The config file token takes precedence over
-
Restart required: Restart Claude Desktop after editing the config file
-
First run: The first time using
npx, it will download and cache the package (may take a few seconds)
✓ Verification: Open Claude Desktop and type: "List my Slack channels". If configured correctly, Claude will use the list_channels tool.
With Other MCP Clients
The server runs on stdio transport and can be integrated with any MCP client:
node build/index.js
Development Mode
For development with hot reload:
npm run dev
Tool Examples
List Channels
{
"name": "list_channels",
"arguments": {
"types": "public_channel,private_channel",
"exclude_archived": true,
"limit": 50
}
}
Send Message
{
"name": "send_message",
"arguments": {
"channel": "C1234567890",
"text": "Hello from MCP!"
}
}
Send Formatted Message with Block Kit
{
"name": "send_formatted_message",
"arguments": {
"channel": "C1234567890",
"text": "Fallback text",
"blocks": [
{
"type": "section",
"text": {
"type": "mrkdwn",
"text": "*Hello* from MCP with _formatting_!"
}
}
]
}
}
Search Messages
{
"name": "search_messages",
"arguments": {
"query": "important announcement",
"count": 10,
"sort": "timestamp"
}
}
Upload File
{
"name": "upload_file",
"arguments": {
"channels": ["C1234567890"],
"content": "File content here",
"filename": "report.txt",
"title": "Monthly Report"
}
}
Add Reaction
{
"name": "add_reaction",
"arguments": {
"channel": "C1234567890",
"timestamp": "1234567890.123456",
"name": "thumbsup"
}
}
Required OAuth Scopes
Make sure your Slack app has these scopes:
| Scope | Purpose |
|---|---|
channels:read |
List and get channel information |
channels:write |
Create channels |
channels:manage |
Archive channels |
users:read |
List and get user information |
chat:write |
Send, update, and delete messages |
files:write |
Upload files |
reactions:write |
Add and remove reactions |
team:read |
Get workspace information |
search:read |
Search messages |
Project Structure
slack-bot-mcp/
├── src/
│ ├── index.ts # MCP server entry point
│ ├── tools/
│ │ ├── channels.ts # Channel management tools
│ │ ├── users.ts # User management tools
│ │ ├── messages.ts # Messaging tools
│ │ ├── files.ts # File upload tool
│ │ ├── reactions.ts # Reaction tools
│ │ └── workspace.ts # Workspace info tool
│ ├── config/
│ │ └── credentials.ts # Environment config loader
│ ├── types/
│ │ └── slack.ts # TypeScript interfaces
│ └── utils/
│ ├── slack-client.ts # Slack API wrapper
│ └── validators.ts # Zod input validators
├── build/ # Compiled JavaScript
├── .env # Your environment variables
├── .env.example # Environment template
├── package.json
├── tsconfig.json
└── README.md
Troubleshooting
Setup Issues
"SLACK_BOT_TOKEN environment variable is required"
Cause: The server can't find your bot token.
Solutions:
- Verify
.envfile exists in project root:ls -la .env - Check token format:
cat .env(should showSLACK_BOT_TOKEN=xoxb-...) - If using Claude Desktop config, ensure the token is in the
envsection - Restart your MCP client after making changes
"Missing required OAuth scope"
Cause: Your bot token doesn't have necessary permissions.
Solution:
- Go to https://api.slack.com/apps → Your App → "OAuth & Permissions"
- Add the missing scope (error message will specify which one)
- Click "Reinstall to Workspace" (this is required after adding scopes)
- Copy the new bot token and update your configuration
💡 Note: You must reinstall the app after changing scopes. The old token won't gain new permissions automatically.
Runtime Issues
"Channel not found" or "User not found"
Cause: You're using channel/user names instead of IDs.
Solution:
- Channel IDs start with
C(public),G(private), orD(DM) - User IDs start with
U - Use
list_channelsorlist_userstools to find the correct IDs
Example:
// ❌ Wrong
{"channel": "general"}
// ✓ Correct
{"channel": "C1234567890"}
"invalid_auth" or "token_revoked"
Cause: Bot token is incorrect or expired.
Solution:
- Go to https://api.slack.com/apps → Your App → "OAuth & Permissions"
- Copy the current "Bot User OAuth Token"
- Update your
.envfile or Claude Desktop config - Restart the MCP server
Rate Limiting
Cause: Slack API has rate limits (Tier 3: 50+ requests per minute).
Solution:
- Add delays between rapid requests
- Use pagination parameters (limit, cursor) for large data sets
- Monitor rate limit headers in error messages
MCP Inspector Won't Start
Cause: Port already in use or Node.js issues.
Solution:
# Kill any existing inspector processes
pkill -f "mcp.*inspector"
# Try again
npx @modelcontextprotocol/inspector node build/index.js
Getting Help
If you're still stuck:
- Check the build output: Run
npm run buildand look for TypeScript errors - Verify Node version: Run
node --version(need 18+) - Test token manually: Use Slack's API tester at https://api.slack.com/methods/auth.test/test
- Enable debug logging: Set
DEBUG=*environment variable
Still having issues? Open an issue at the GitHub repository with:
- Error message (full text)
- Your Node.js version
- Steps to reproduce
Development
Building
npm run build
Running in Development
npm run dev
Testing with MCP Inspector
npx @modelcontextprotocol/inspector node build/index.js
License
ISC
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
Acknowledgments
Ported from the Python implementation: https://github.com/piekstra/slack-mcp-server
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.