MCP Server

CommunicationTeam WorkflowMIT License

Slack MCP Server

Q: What is the Slack MCP Server?

Slack MCP Server is a Model Context Protocol integration that connects AI coding assistants to Slack workspaces. It lets Claude Code, Cursor, and other MCP clients send messages, search channels, read thread history, list channels, and manage Slack interactions through natural language — without switching to the Slack app.

Q: How do I set up Slack MCP Server?

Create a Slack App at api.slack.com/apps with the required OAuth scopes (channels:read, channels:history, chat:write, users:read). Install it to your workspace and copy the Bot User OAuth Token. Then configure the MCP server with: claude mcp add slack -- npx @anthropic/mcp-server-slack, setting the SLACK_BOT_TOKEN environment variable.

Q: What Slack permissions does the MCP server need?

At minimum: channels:read (list channels), channels:history (read messages), chat:write (send messages), users:read (resolve usernames). For DMs, add im:read and im:write. For thread replies, channels:history is sufficient. The principle of least privilege applies — only grant scopes your workflow needs.

Q: Can the Slack MCP Server read private channels?

Yes, if you add the groups:read and groups:history scopes and invite the bot to the private channel. The MCP server respects Slack API permissions — it can only access channels and messages that the bot token has been granted access to.

Q: Is it safe to let an AI assistant post to Slack?

The MCP server requires explicit confirmation before posting messages in most clients. Claude Code shows the message content and target channel before sending. For safety, start with read-only scopes (channels:read, channels:history) and only add chat:write after you are comfortable with the integration.

Q: What are common use cases for Slack MCP?

Top use cases: posting deployment notifications from CI/CD workflows, searching for prior decisions in channel history, summarizing long threads, posting standup updates from your development session, and routing alerts from monitoring tools. The agent can also search across channels to find relevant conversations before you interrupt teammates.

by Anthropic · api.slack.com

Slack MCP Server bridges your AI coding assistant directly into your team communication. Instead of context-switching between your IDE and Slack, your agent can search channel history for prior decisions, post deployment updates, summarize long threads, and route alerts — all through natural language in your coding session.

For development teams, this means fewer interruptions. The agent can check if someone already answered a question in Slack before you ask it again, post PR links with context, and send standup updates without opening the Slack app.

Capabilities

send, search, read...

<5min

Setup

Slack App + npx

750K+

Workspaces

Slack users

OAuth

Security

scoped permissions

Quick Install

claude mcp add slack -- npx @anthropic/mcp-server-slack

Key Features

Send Messages

Post messages to any channel or DM. Include formatted text, code blocks, and links. The agent previews the message before sending for your approval.

Search Channels

Search across channels for keywords, past decisions, or specific conversations. Find answers without scrolling through hundreds of messages manually.

Read Thread History

Read full thread conversations including replies. The agent can summarize long threads into actionable bullet points in seconds.

List Channels & Users

List all channels in the workspace, check membership, and resolve usernames. Useful for routing messages to the right channel.

Post Deployment Updates

Automatically post deployment notifications with commit details, PR links, and status to your team channel after successful deploys.

Standup Automation

Generate and post standup updates based on your recent git commits, open PRs, and current work context — without manual writing.

Execution Brief

Use this page as a rollout checklist, not just reference text.

Suggest update

Tool Mapping Lens

Organize Tools by Workflow Phase

Catalog-oriented pages work best when users can map discovery, evaluation, and rollout in a clear path instead of reading an undifferentiated list.

Define the job-to-be-done first
Group tools by stage
Prioritize by adoption friction

Actionable Utility Module

Skill Implementation Board

Use this board for Slack MCP Server before rollout. Capture inputs, apply one decision rule, execute the checklist, and log outcome.

Input: Objective

Deliver one measurable improvement with slack mcp server claude code setup team communication review

Input: Baseline Window

20-30 minutes

Input: Fallback Window

8-12 minutes

Decision Trigger	Action	Expected Output
Input: one workflow objective and release owner are defined	Run preview execution with fixed acceptance criteria.	Go or hold decision backed by repeatable evidence.
Input: output quality below baseline or retries increase	Limit scope, isolate root issue, and rerun controlled test.	One confirmed correction path before wider rollout.
Input: checks pass for two consecutive replay windows	Promote to broader traffic with fallback path active.	Stable rollout with low operational surprise.

Execution Steps

Record objective, owner, and stop condition.
Execute one controlled preview run.
Measure quality, latency, and correction burden.
Promote only when pass criteria are stable.

Output Template

tool=slack mcp server claude code setup team communication review
objective=
preview_result=pass|fail
primary_metric=
next_step=rollout|patch|hold

Share execution feedback

What Is Slack MCP Server?

Slack MCP Server is a Model Context Protocol integration that connects AI coding assistants to Slack workspaces. It exposes Slack API operations as MCP tools, enabling your agent to send messages, search conversations, read threads, list channels, and manage team communication without leaving your development environment.

The server authenticates through a Slack Bot Token with configurable OAuth scopes, giving you precise control over what the AI can access. Start with read-only scopes to search and summarize, then add write scopes when you trust the workflow. Every write operation requires explicit confirmation in your MCP client.

For engineering teams, the Slack MCP Server eliminates a major source of context switching. Instead of tabbing to Slack to check if a question was already answered, your agent searches channel history instantly. Instead of manually posting deployment updates, the agent generates and posts them with relevant context from your git history.

The integration is especially powerful when combined with other MCP servers. GitHub MCP + Slack MCP means your agent can open a PR, then post a summary to the team channel. Brave Search + Slack means the agent can research an answer, then share it in the relevant thread. The composability of MCP servers makes each one more valuable.

How to Calculate Better Results with slack mcp server claude code setup team communication review

Create a Slack App at api.slack.com/apps. Add Bot Token Scopes: channels:read, channels:history, chat:write, users:read. Install the app to your workspace and copy the Bot User OAuth Token.

Install the MCP server: claude mcp add slack -- npx @anthropic/mcp-server-slack. Set SLACK_BOT_TOKEN as an environment variable.

Invite the bot to channels you want it to access: /invite @your-bot-name in each Slack channel. The bot can only read channels it has been invited to.

Test by asking your agent to list channels or search for a recent message. Start with read operations before testing writes to build confidence in the integration.

Treat this page as a decision map. Build a shortlist fast, then run a focused second pass for security, ownership, and operational fit.

When a team keeps one shared selection rubric, tool adoption speeds up because evaluators stop debating criteria every time a new option appears.

Worked Examples

Automated deployment notification workflow

You finish a feature and deploy to production via Claude Code
Ask: "Post a deployment update to #engineering with what we just shipped"
Agent reads recent git commits and the PR description via GitHub MCP
Agent drafts a formatted Slack message with feature summary, PR link, and deploy timestamp
Agent shows you the preview and asks for confirmation
You approve — agent posts to #engineering via Slack MCP
Team sees the update immediately with full context

Outcome: A deployment notification with real context posted in 30 seconds instead of a manual copy-paste job that takes 5 minutes and often gets skipped.

Finding prior decisions before asking the team

You need to know which auth library the team chose last month
Ask: "Search Slack for discussions about auth library selection in #architecture"
Agent searches channel history via Slack MCP and finds the thread
Agent summarizes: "The team chose Clerk over Auth0 in a March 3 thread. Reasons: better DX, lower pricing, built-in user management. Decision by @sarah."
You proceed with Clerk integration without interrupting anyone
No Slack notification was sent — pure read operation

Outcome: Found a prior team decision in 15 seconds through AI-assisted search instead of asking in the channel and waiting for someone to remember.

Frequently Asked Questions

What is the Slack MCP Server?