MCP ConfigsApr 6, 2026·2 min read

Slack MCP — Team Messaging Server for AI Agents

MCP server that connects AI agents to Slack workspaces. Send messages, read channels, search history, manage threads, and automate team communication via natural language. 1,500+ stars.

TL;DR
Slack MCP lets AI agents send messages, read channels, and search Slack workspaces.
§01

What it is

Slack MCP is a Model Context Protocol server that connects AI agents to Slack workspaces. Through this integration, AI assistants like Claude Code or Cursor can send messages, read channels, search conversation history, manage threads, and automate team communication using natural language commands.

This server targets development teams who want their AI coding tools to interact with Slack without manual copy-paste. Instead of switching to Slack to post an update, you tell your AI agent to send a message to a specific channel.

§02

Why it saves time or tokens

Communicating deployment status, code review summaries, or incident updates to Slack channels typically requires switching context. Slack MCP eliminates this by letting the agent post directly from the coding environment. Searching Slack history for past decisions or context is also faster when the agent does it inline rather than requiring you to open the Slack app and search manually.

§03

How to use

  1. Install the Slack MCP server in your AI tool's MCP configuration
  2. Configure a Slack Bot Token with the necessary OAuth scopes
  3. Ask your AI agent to read channels, post messages, or search conversations
§04

Example

{
  'mcpServers': {
    'slack': {
      'command': 'npx',
      'args': ['-y', '@anthropic/slack-mcp-server'],
      'env': {
        'SLACK_BOT_TOKEN': 'xoxb-your-token-here'
      }
    }
  }
}

With this config, your AI agent can execute commands like 'post to #deployments: v2.3.1 deployed to production, all health checks passing.'

ActionDescription
Send messagePost to any channel the bot is in
Read channelFetch recent messages from a channel
SearchSearch message history by keyword
Thread replyReply to a specific message thread
List channelsDiscover available channels
§05

Related on TokRepo

§06

Common pitfalls

  • The Slack Bot Token requires specific OAuth scopes (channels:read, chat:write, search:read); missing scopes cause silent permission errors
  • The bot must be invited to each channel before it can read or post; it cannot access channels it has not joined
  • Rate limits on the Slack API can throttle rapid-fire operations; batch messages where possible

Frequently Asked Questions

What OAuth scopes does the Slack bot need?+

At minimum, the bot needs channels:read, chat:write, and users:read scopes. For searching message history, add search:read. For managing threads, add channels:history. For posting in private channels, add groups:write. Start with minimal scopes and add more as needed.

Can Slack MCP read private channels?+

Yes, if the bot has the groups:read and groups:history scopes and has been invited to the private channel. The bot cannot discover or access private channels it has not been explicitly added to. An admin or channel member must invite the bot.

Is the bot token stored securely?+

The token is stored in your local MCP configuration file. It never leaves your machine unless sent to Slack's API over HTTPS. Treat it like any API secret: do not commit it to version control, and use environment variables or a secrets manager in shared environments.

Can I use Slack MCP with Claude Desktop?+

Yes. Claude Desktop supports MCP servers through its configuration file. Add the Slack MCP server entry to your Claude Desktop MCP settings, provide the bot token, and Claude can interact with your Slack workspace directly from the desktop app.

How does Slack MCP handle rate limits?+

Slack MCP passes API responses back to the AI agent, including rate limit headers. If a rate limit is hit, the agent receives an error and can retry after the cooldown period. For bulk operations, the agent should pace requests to stay within Slack's tier-based rate limits.

Citations (3)
🙏

Source & Thanks

Built on Slack API. Licensed under MIT.

mcp-slack — ⭐ 1,500+

Thanks for making team communication AI-accessible.

Discussion

Sign in to join the discussion.
No comments yet. Be the first to share your thoughts.