Slack MCP

Overview

Slack MCP is a Model Context Protocol server that enables AI agents to interact with Slack workspaces. It allows agents to read messages, send messages, manage channels, and integrate with Slack's collaboration features.

Key Features

💬 Messaging

Read Messages: Fetch messages from channels and DMs
Send Messages: Post messages to channels or users
Reply to Threads: Respond to specific thread messages
React to Messages: Add emojis to messages

📺 Channel Management

List Channels: View all public and private channels
Create Channels: Create new channels
Join/Leave Channels: Manage channel membership
Get Channel Info: View channel details and members

👥 User Management

List Users: View all workspace members
Get User Info: View user profile details
Get User Email: Retrieve user email addresses

🔍 Search

Search Messages: Find messages by keyword
Search Channels: Find channels by name
Filter by Date: Search within date ranges

Installation

# Using npx (recommended)
npx -y @modelcontextprotocol/server-slack

# With Slack token
SLACK_BOT_TOKEN=xoxb-your-bot-token npx -y @modelcontextprotocol/server-slack

Configuration

Required Setup

Create a Slack App:
- Visit Slack API
- Click "Create New App" → "From scratch"
- Name your app and select your workspace
Configure OAuth Scopes: Go to OAuth & Permissions and add these scopes:
- channels:history - View messages in channels
- channels:read - View channel list and info
- channels:manage - Create and manage channels
- chat:write - Send messages
- users:read - View user list and info
- users:read.email - View user emails
- reactions:write - Add emoji reactions
- search:read - Search messages and files
Install to Workspace:
- Click "Install to Workspace"
- Authorize the app
- Copy the Bot User OAuth Token (starts with xoxb-)
Enable Message Subscription (optional, for real-time updates):
- Go to Event Subscriptions
- Enable events
- Add bot event subscriptions

Configure in Claude Desktop:

{
  "mcpServers": {
    "slack": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-slack"],
      "env": {
        "SLACK_BOT_TOKEN": "xoxb-your-bot-token-here"
      }
    }
  }
}

Available Tools

Tool	Description
`list_channels`	List all accessible channels
`get_channel_info`	Get details about a specific channel
`get_channel_members`	Get members of a specific channel
`join_channel`	Join a channel
`leave_channel`	Leave a channel
`create_channel`	Create a new channel
`list_users`	List all workspace users
`get_user_info`	Get details about a specific user
`get_user_email`	Get email address of a user
`get_messages`	Get messages from a channel
`post_message`	Post a message to a channel
`reply_to_thread`	Reply to a message thread
`add_reaction`	Add an emoji reaction to a message
`search_messages`	Search for messages by keyword
`search_channels`	Search for channels by name

Usage Examples

List Channels

Show me all the channels I'm in

Returns:

Channels:
- #general (156 members)
- #engineering (89 members)
- #product (45 members)
- #random (203 members)
- #announcements (156 members)

Get Channel Messages

Show me the last 10 messages from #engineering

Returns recent messages with timestamps and authors.

Search Messages

Find messages mentioning "deployment" from last week

Agent searches with:

Query: deployment
Date range: Last 7 days

Post a Message

Post this message to #general: "Team meeting tomorrow at 2pm"

Create a Channel

Create a new channel called #project-alpha

Get User Info

Who is @john.doe?

Returns:

Name: John Doe
Email: john.doe@company.com
Title: Senior Engineer
Status: 🚀 Shipping v2.0
Timezone: America/Los_Angeles

Reply to Thread

Reply to the last message in #engineering with "Thanks for the update!"

Add Reaction

Add a 👍 reaction to the message "Deployment successful" in #engineering

Search for Files

Find PDF files shared in #product last month

Integration with AI Agents

Meeting Assistant

Claude: "I see you have a team meeting in #general..."
[Reads meeting announcement]
Claude: "Would you like me to create a follow-up channel for meeting notes?"
[Creates #meeting-notes-2026-05-15]
Claude: "I'll track action items from the meeting and post summaries..."

Customer Support

User: "Someone asked about pricing in #support"
Claude: "Let me find that message..."
[Searches #support for pricing questions]
Claude: "Here's the message. Would you like me to draft a response?"
[Drafts response, posts to thread]

Team Coordination

Claude: "I noticed @alice mentioned a blocker in #engineering..."
[Reads message, identifies blocker]
Claude: "Should I notify the engineering lead?"
[Posts notification to #engineering-lead]

Daily Digest

Claude: "Here's your daily Slack digest..."
[Summarizes mentions, reactions, and important messages]
Claude: "You have 3 unread messages in #project-alpha and 2 mentions..."

Custom Agent Integration

from mcp import ClientSession, StdioServerParameters
from mcp.client.stdio import stdio_client
import os

# Connect to Slack MCP
server_params = StdioServerParameters(
    command="npx",
    args=["-y", "@modelcontextprotocol/server-slack"],
    env={"SLACK_BOT_TOKEN": os.environ["SLACK_BOT_TOKEN"]}
)

async with stdio_client(server_params) as (read, write):
    async with ClientSession(read, write) as session:
        await session.initialize()
        
        # List channels
        channels = await session.call_tool("list_channels", arguments={})
        
        # Get messages
        messages = await session.call_tool(
            "get_messages",
            arguments={"channel_id": "C123456", "limit": 10}
        )
        
        # Post message
        result = await session.call_tool(
            "post_message",
            arguments={
                "channel_id": "C123456",
                "text": "Hello from the AI agent!"
            }
        )
        
        # Search messages
        results = await session.call_tool(
            "search_messages",
            arguments={"query": "deployment", "count": 20}
        )

Security Considerations

Bot Token Security

⚠️ Important: Protect your Slack bot token:

Never commit tokens to version control
Use environment variables or secret managers
Rotate tokens regularly
Use the minimum required scopes

Channel Access

Bot only sees channels it's invited to
Private channels require explicit invitation
Consider using a dedicated bot user for MCP

Message Privacy

Bot can only read messages in channels it has access to
DMs require additional permissions
Be mindful of sensitive information in channels

Pros

✅ Real-Time Collaboration: Direct Slack integration
✅ Rich Messaging: Send, read, reply, react
✅ Channel Management: Create and manage channels
✅ User Directory: Access workspace user information
✅ Search Capabilities: Find messages and channels
✅ Thread Support: Work with threaded conversations

Cons

❌ Token Required: Must create Slack app and bot
❌ Scope Permissions: Need multiple OAuth scopes
❌ Rate Limits: Slack API has rate limits
❌ Bot Limitations: Can't access all user features
❌ Setup Complexity: Multi-step Slack app setup

When to Use

Choose Slack MCP when:

Your team uses Slack for communication
You want AI to participate in Slack conversations
You need to search and analyze Slack messages
You want automated Slack notifications

Consider alternatives when:

Your team uses Discord, Teams, or other platforms
You only need occasional Slack access
You need advanced Slack workflow automation

Resources

GitHub: https://github.com/modelcontextprotocol/servers/tree/main/src/slack
Slack API Docs: https://api.slack.com/
Slack App Management: https://api.slack.com/apps
MCP Servers Repo: https://github.com/modelcontextprotocol/servers

Comparison

Feature	Slack MCP	Discord MCP	Teams MCP
Platform	Slack	Discord	Microsoft Teams
Setup	⚠️ Moderate	✅ Easy	⚠️ Complex
Message Access	✅ Full	✅ Full	⚠️ Limited
Channel Mgmt	✅ Yes	✅ Yes	⚠️ Limited
Search	✅ Yes	✅ Yes	⚠️ Limited
Enterprise	✅ Yes	⚠️ No	✅ Yes

Last updated: May 2026

Slack MCP

Overview

Setup

Configuration

Documentation

Slack MCP

Overview

Key Features

💬 Messaging

📺 Channel Management

👥 User Management

🔍 Search

Installation

Configuration

Required Setup

Available Tools

Usage Examples

List Channels

Get Channel Messages

Search Messages

Post a Message

Create a Channel

Get User Info

Reply to Thread

Add Reaction

Search for Files

Integration with AI Agents

Meeting Assistant

Customer Support

Team Coordination

Daily Digest

Custom Agent Integration

Security Considerations

Bot Token Security

Channel Access

Message Privacy

Pros

Cons

When to Use

Resources

Comparison