slack-mcp-server

What is slack-mcp-server?

An MCP (Model Context Protocol) server that provides a standardized interface for AI assistants to interact with the Slack API. It allows reading and writing Slack data through configurable bot and user tokens.

How to use slack-mcp-server?

Install via npm install @ubie-oss/slack-mcp-server, set environment variables SLACK_BOT_TOKEN and SLACK_USER_TOKEN (and optionally SLACK_SAFE_SEARCH), then run with npx @ubie-oss/slack-mcp-server (default stdio) or with -port 3000 for Streamable HTTP. Configure clients with the provided JSON block.

Key features of slack-mcp-server?

• List public channels with pagination
• Post messages and reply to threads
• Add emoji reactions to messages
• Retrieve channel history and thread replies
• Search messages with powerful filters
• Get user profiles in bulk

Use cases of slack-mcp-server?

• Automate sending notifications or updates to Slack channels
• Search workspace messages using date, user, and content filters
• Manage thread discussions by posting replies and reading history
• Fetch user profile information in batch for AI-driven workflows
• Enable AI assistants to interact with Slack via MCP-compatible clients

FAQ from slack-mcp-server

What tokens are required?

SLACK_BOT_TOKEN (bot user OAuth token) is required for most operations. SLACK_USER_TOKEN (user OAuth token) is needed for features like message search.

What does SLACK_SAFE_SEARCH do?

When set to true, it automatically excludes private channels, DMs, and group DMs from search results. This is enforced server-side and cannot be overridden by clients.

What transports are supported?

Stdio transport (default for local integration) and Streamable HTTP transport (specify with -port 3000 for web applications). The HTTP endpoint is http://localhost:3000/mcp.

Where is this package hosted and how do I authenticate?

The package is hosted on GitHub Registry.