Live today! Discover, install, and trust MCP servers with one click
guidesUpdated February 26, 2026

Slack MCP server: let your AI send messages, search history, and more

The Slack MCP server connects your AI assistant to Slack. Search messages, send updates, list channels, and get channel history without switching apps.

What the Slack MCP server does

Slack is where work happens in real time. The Slack MCP server connects your AI assistant to your workspace so it can participate in that workflow.

With the Slack MCP server connected:

  • "Send a message to #engineering: the deploy is live"
  • "What did people say in #product this morning?"
  • "List all channels I am a member of"
  • "Find the thread where we discussed the API rate limits"
  • "Post a summary of today's incidents to #on-call"

Your AI can act as a Slack operator, pulling information out of your channels and putting summaries back in, without you opening the app.

New to MCP? Read What is an MCP server? first, or see How to install an MCP server for a general installation guide.

What it exposes

AI ClientClaude, Cursor, Windsurf+ any MCP-compatible appMCPprotocolMCP ServerSlack adaptertranslates AI calls to APISlackAPISlackyour data and actions

The Slack MCP server wraps the Slack Web API and exposes:

Messages: send to channels, post to threads, send direct messages

Channels: list, search, get history, get members

Search: full-text search across messages and files

Users: get user info, list workspace members

Prerequisites

  • A Slack workspace where you have admin access or permission to install apps
  • A Slack Bot Token (OAuth)
  • Node.js 18+ and an MCP-compatible client

Setup

Step 1: Create a Slack app

Go to api.slack.com/apps and create a new app from scratch. Under "OAuth and Permissions", add the scopes your use case requires:

  • Read: channels:history, channels:read, groups:history, im:history, search:read, users:read
  • Write: chat:write, chat:write.public

Install the app to your workspace and copy the Bot User OAuth Token.

Step 2: Configure your MCP client

{
  "mcpServers": {
    "slack": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-slack"],
      "env": {
        "SLACK_BOT_TOKEN": "xoxb-your-token-here",
        "SLACK_TEAM_ID": "T0123456789"
      }
    }
  }
}

What to watch out for

Write permissions are powerful. Before granting chat:write, make sure you are comfortable with the AI sending messages on your behalf. Start read-only and expand permissions as you get comfortable.

Rate limits. Slack's free tier has search rate limits. If you query history frequently, you may hit them. The MCP server will surface errors when this happens.

Private channels. Your bot can only read channels it has been added to. Use /invite @your-bot-name in any private channel you want the AI to access.

Local setup

The setup above runs the Slack server locally via npx. This means Node.js or Python needs to be installed on your machine. Check the MCP Marketplace listing to compare the available Slack servers and find the best fit for your workflow.

Find it on MCP Marketplace

Browse the Slack MCP server on MCP Marketplace with security scores and ratings, or explore the full list of communication and collaboration servers.

For a curated list of other useful developer servers, see Best MCP servers for developers.

Browse MCP servers

Find the servers mentioned in this post and thousands more on MCP Marketplace. Security-checked, one-click install.

Browse servers

Keep reading