Skip to main content

Slack MCP server overview

Model Context Protocol (MCP) is a protocol that lets AI clients securely access data sources. MCP servers implement that protocol for specific tools. The Slack MCP server provides secure access to your Slack workspace, allowing AI agents and other integrations to interact with your team's data.

With the Slack MCP server, your integrations can search channels, send messages, and perform other Slack actions through MCP-compatible clients. Workspace admins can approve and manage all MCP client integrations, keeping your Slack data safe.

The Slack MCP server is currently being rolled out to select partners.

If you are a partner interested in participation, please reach out to your Slack partner manager. For everyone else, here's a sneak peek at what's to come.

Available tools​

The Slack MCP server provides tools for:

Each of these tools provides useful functionality for interacting with Slack; combine them for comprehensive integrations that grasp your team's context and history.

Searching throughout Slack​

The MCP server can search for a variety of information found throughout a Slack workspace:

  • Search messages and files - filter by date, user, and content type. Retrieve metadata and content.
  • Search users in a workspace - filter by name (with partial name matching), email, and user ID. Retrieve user details and statuses.
  • Search private and public channels - filter by channel name and description! Retrieve channel metadata.

Retrieving and sending messages​

The MCP server can retrieve and send messages throughout a Slack workspace:

  • Send messages - send messages to any type of conversation in Slack.
  • Read channels - grab the complete message history of channels.
  • Read threads - grab complete message thread conversations.

Managing canvases​

The MCP server can interact and modify Slack canvases:

  • Create a canvas - create and share rich, formatted documents.
  • Read a canvas - export canvases as markdown files.

Managing users​

The MCP server can also fetch user info:

  • Fetch user profiles - access complete user profile info, including custom profile fields and statuses.

Example use cases​

There are quite a lot of possibilities with the Slack MCP server. Here are just a few ideas:

  • Create an AI assistant in Slack that can search through your team's Slack history to answer questions, find past decisions, and provide context for current projects.
  • Bring content from outside Slack into Slack via messages and canvases for discussion with coworkers.
  • Bring content from Slack to AI agents, providing them full context of projects that exist across multiple products.

Available clients​

The MCP server is available to select partners in two clients:

Stay tuned for more info as we build out this feature.