Model Context Protocol

Slack MCP Server

Communication & Messaging via Model Context Protocol

Connect your AI agents to Slack channels. Read messages, post updates, and manage team communication through a standardized protocol.

Operations and engineering teams use Slack MCP to make their Slack workspace intelligent and actionable. Instead of building and maintaining custom Slack apps with bolt.js, your AI agent can read channels, understand context, and post meaningful updates—all through a standard protocol. This means your team gets AI-powered notifications, automated standups, and smart alerting directly in the channels they already use.

  • ✓AI-driven notifications and alerts without custom Slack apps
  • ✓Read channel context to inform AI decisions
  • ✓Automate status updates and reporting
  • ✓Bridge conversations between Slack and other tools
Deploy Managed ServerExplore Integrations
mcp-config.json
{
  "mcpServers": {
    "slack": {
      "command": "sse",
      "args": [
  "https://slack-mcp.boltai.com"
],
      "env": {
        
      }
    }
  }
}
Real-World Automation

Common Workflows

See how teams combine this MCP with other tools to automate real business processes.

Intelligent Incident Alerts

Scenario: Your monitoring system detects an outage. AI analyzes severity and composes a clear, actionable Slack message to #incidents.
Steps:
  1. Firecrawl or metric DB indicates error spike >5σ
  2. AI consults runbook (from Notion) to determine severity
  3. Slack MCP posts to #incidents with structured details
  4. Mentions @on-call-engineer and includes runbook link
  5. Linear MCP creates incident ticket automatically
Outcome: MTTD (mean time to detect) drops from 15 min to <30 sec. On-call engineers alerted instantly with context, not just an alarm.

Daily Standup Bot

Scenario: 9 AM every weekday. AI compiles yesterday's wins, today's plan, and blockers from multiple sources and posts to #standup.
Steps:
  1. Linear MCP fetches yesterday's completed tasks for team
  2. Calendar MCP (if available) reads today's meetings
  3. Notion MCP checks OKR progress updates
  4. AI composes natural language summary with highlights
  5. Slack MCP posts formatted message to #team-standup
Outcome: Team saves 20 minutes daily on manual updates. Leadership gets consistent visibility without chasing.

Deployment Notifier

Scenario: CI/CD pipeline finishes. AI announces the deployment, notes changes, and links to changelog and monitoring.
Steps:
  1. GitHub/GitLab webhook triggers n8n or direct
  2. AI reads commit messages and PR tickets (from Linear/GitHub)
  3. Summarizes changes in plain language for non-engineers
  4. Slack MCP posts to #releases with rollback instructions if needed
  5. Pin message for 24 hours for visibility
Outcome: Company-wide communication of releases automated. Support and sales can anticipate changes. Reduction in 'what changed?' queries by 90%.
Protocol Definition

Available Tools — In Depth

Detailed reference for each tool exposed by this MCP server, with examples and related use cases.

⚒

post_message

Send a text or formatted message to any Slack channel or DM. Supports Slack's mrkdwn formatting (bold, italic, links, code blocks, emojis), threads (reply in thread), and mentions (@username, @channel, @here). Essential for AI-to-team communication. You can also include blocks for rich layouts (buttons, images).

Example:""Post to #sales-leads: '🚀 New high-value lead from Acme ($50k ARR potential). Owner: @john. CRM link: hubspot.com/deals/123' with :fire: emoji.""
Works great with:read_historylist_channels
⚒

list_channels

List all accessible Slack channels (public and private) with IDs, names, and member counts. Use this to discover where you can post messages or read history. Returns only channels your bot has been invited to.

Example:""List all channels with 'alert' or 'monitor' in the name. Return channel IDs and purpose.""
Works great with:post_messageread_history
⚒

read_history

Fetch recent messages from a channel (up to 1,000 messages). Returns message text, author, timestamp, and thread info. Use this to summarize channel discussions or to check if something was already announced before posting.

Example:""Read last 50 messages from #incidents today. Summarize any outages or deployments mentioned.""
Works great with:post_messagelist_channels
Setup Guide

Configuration & Best Practices

Setup Checklist

  • ✓
    Slack workspace & bot setup
    Create a Slack app at api.slack.com/apps. Enable 'Bot Token Scopes': chat:write, channels:read, groups:read, im:read, mpim:read, users:read.
  • ✓
    OAuth token
    Install app to workspace → copy 'xoxb-' Bot User OAuth Token. Never share publicly.
  • ✓
    Invite bot to channels
    Bot can only read channels it's a member of. Invite manually: /invite @YourBot in each channel.
  • ✓
    Rate limits
    Slack API: ~50-100 messages/min per workspace. Exceeding returns rate_limited error with Retry-After header.
  • ✓
    Mentions formatting
    Use <@U12345> for user, <@C12345> for channel, <!channel> for @channel. The MCP may auto-format usernames.

Troubleshooting

403 not_in_channel
Fix: Bot hasn't joined the channel. Invite it first using /invite @bot or have an admin invite.
404 channel_not_found
Fix: Channel name/ID typo OR bot not a member of private channel. Check ID via list_channels.
invalid_auth / token_revoked
Fix: Token invalid, expired, or app uninstalled. Regenerate token in Slack API dashboard.
rate_limited
Fix: Hit Slack burst limit. Wait Retry-After seconds before retrying. Implement backoff.
Rate limit: ~50-100 messages/minute depending on workspace plan. Read APIs: ~100 requests/minute. Burst limits apply.

When to Use Slack MCP Server vs. Alternatives

Use This MCP When:

  • You need AI-native access via natural language
  • Your workflows span multiple tools (MCP composability)
  • You prefer cloud hosting over local Docker
  • You want zero-config deployment with ClawFast
  • Your use case requires LLMs to reason and act autonomously

Consider Alternatives When:

  • You need bulk data sync (use native export/import)
  • Real-time streaming is critical (use native webhooks)
  • You have strict compliance requiring direct API audit logs
  • Your integration is a one-off script (direct SDK may be simpler)
  • You need features not yet exposed by this MCP server
FAQ

Common Questions About Slack MCP Server

Q: Do users see the bot as a real user?
A: Bot posts appear as 'Your Bot Name' with a bot icon. Can be customized in Slack app settings.
Q: Can the bot read DMs?
A: Yes, but only DMs where the bot is a participant. Users must DM the bot first or be in a multi-party DM with it.
Q: Can it post in channels it's not in?
A: No. Bot must be invited to every channel where it will post or read. Public channels require invitation; private channels need member invite.
Q: Thread support?
A: post_message can reply in a thread by providing thread_ts (timestamp). Use read_history to get thread_ts from parent message.
Q: File uploads?
A: Not supported by this MCP—only text messages. Upload files using Slack's files.upload API via a separate connector or n8n.
Q: Markdown vs mrkdwn?
A: Slack uses mrkdwn (similar but not identical to Markdown). Formatting: *bold*, _italic_, `code`, > blockquote. Links: <http://url|text>.
Q: Can it mention @here or @channel?
A: Yes—include <!channel> or <!here> in the message text. Some workspaces restrict these mentions for bots.
Q: Message history limit?
A: Default returns up to 1,000 messages. For larger archives, paginate using cursor parameter (not yet in basic MCP).
Customer Success

Success Story: RapidScale (DevOps agency)

Challenge

Engineers missed critical alerts buried in email. On-call confusion led to 30+ minute response times during outages. Stakeholders had no visibility into incident progress.

Solution

Deployed Slack MCP + Firecrawl + Linear MCP. AI monitors error logs, assesses severity, posts structured incident alerts to #incidents with runbook links, creates Linear tickets, and posts resolution summaries.

Results

  • ✓ MTTR (time to resolution) improved 58%: 45 min → 19 min
  • ✓ False positive alerts dropped 70% (AI filters noise)
  • ✓ Stakeholder inquiries eliminated—everyone sees #incidents updates live
  • ✓ On-call engineer satisfaction score +40% (less chaos)
"Before, our Slack was noise. Now it's our nerve center. Every alert has context, every resolution is documented automatically. We've turned chaos into calm."
— David Kumar, Head of SRE, RapidScale

Combine with Other MCPs

This MCP works great alongside other tools. Here are popular combinations.

linear
Turn Slack alerts into Linear tickets automatically
→
firecrawl
Scrape error dashboards and post summaries to Slack
→
notion
Link incident post-mortems from Slack to Notion runbooks
→
hubspot
Notify sales of customer-reported issues from Slack
→

Ready for full workflows?

Check out our integration page for Slack MCP Server to see complete AI agent templates and step-by-step guides.

View Slack MCP Server Integrations →

Technical Specifications

Protocol
SSE (Server-Sent Events) via Bolt AI hosted instance
Transport
HTTPS (wss for real-time events)
Authentication
Bearer token (Slack Bot User OAuth Token xoxb-*)
Rate Limit
~50-100 msgs/min (workspace-dependent)
Typical Latency
200-600ms (Slack API)
Data Format
JSON with Slack's message block format
Supported LLMs
Claude 3.5+, GPT-4o, Kimi K2.5
Hosting
Bolt AI hosted SSE endpoint (bridges to your Slack)
Uptime SLA
Depends on Bolt AI availability (not ClawFast)
Last Updated
Managed by Bolt AI service

Why Managed MCP?

Model Context Protocol (MCP) is the new standard for AI connectivity. While you can host servers locally, ClawFast provides a production-grade managed environment.

  • 24/7 Availability: No need to keep your local machine running.
  • Secure Secrets: API keys are encrypted at rest and never exposed to logs.
  • SSE Protocol: Native Server-Sent Events (SSE) support for cloud-to-cloud connectivity.
  • Zero Config: One-click deployment with pre-configured templates.

Looking for more ways to use Slack MCP Server?

Explore our high-level integration page for Slack MCP Server to see business use cases and ready-to-use AI agent templates.

View Slack MCP Server Integrations →

Connect Slack MCP Server to your AI stack today

Deploy your managed MCP server in under 60 seconds.

Start Free Deployment