Guide to the Slack MCP server

The Slack Model Context Protocol (MCP) server lets third-party AI assistants like Claude and Perplexity securely access your Slack content so they can search messages, find information, and take actions on your behalf.

Note: This is a general overview of the Slack MCP server. For more detailed information as well as guidance on how to develop your own MCP client, visit the Slack MCP server overview in our developer docs. 

How it works

• Model Context Protocol is an open-source protocol that lets AI clients securely access external data sources. The Slack MCP server provides AI assistants with secure access to your Slack workspace, allowing them to interact with your organization’s data.
• To get started, choose from our partner apps and follow their set up instructions to install and connect them to the Slack MCP server. 
• Once connected, third-party AI assistants can search across your workspace or Enterprise organization, read specific channels or threads, send messages, create canvases, and help you find information faster — all using natural language requests.

What AI assistants can do with the MCP server

The MCP server provides tools for the following actions:

• Search throughout Slack
  Search for messages, files, members, and channels. 
• Retrieve and send messages  
  Read and retrieve channel history and send messages to any type of conversation in Slack. 
• Manage canvases 
  Create and read Slack canvases.
• Access member info 
  Access and read profile information including custom profile fields and statuses. 

Connect with a third-party AI assistant

Get started using the Slack MCP server by setting up a connection with an available partner:

• Claude
• Credal AI
• Cursor
• Dropbox
• Guru
• Jasper

• Notion
• Open AI
• Perplexity
• Thoughtspot
• Wordsmith
• Workleap

Additional partners coming soon!