You can connect Slack to Claude using the Model Context Protocol (MCP), and once it’s connected, Claude can search a channel, summarize a thread, or post an update without you copying and pasting out of Slack by hand. This guide covers what the Slack MCP server does, both ways to connect it — directly or through a gateway — and one thing worth knowing before you try: Slack’s OAuth isn’t currently open to every client that asks.

What the Slack MCP server actually does

Slack publishes a remote MCP server at mcp.slack.com. It exposes your workspace’s messages and channels as tools Claude can call directly, so prompts like these work against your real Slack data:

  • “Summarize #support since yesterday.”
  • “Find the thread where we decided on the pricing change.”
  • “What did I miss while I was out?”
  • “Post these release notes to #general.”

Claude reads and, depending on the tools you allow, writes through the same permissions your Slack account already has — it isn’t a separate export or a bot with its own limited view, it’s your account acting on your behalf.

The catch: Slack’s OAuth isn’t open to every client yet

Most servers in gate’s directory — Notion, Linear, Sentry — let any AI client register itself and send you straight through a one-click OAuth consent screen. Slack’s MCP endpoint is one of a handful where that isn’t the case today: the vendor’s OAuth currently only admits clients Slack has allow-listed in advance, rather than opening registration to any client that shows up. In practice, that can mean the connection attempt from a new or less-common client fails at the authorization step, even though the server itself is real and documented.

This is the same gap covered in MCP OAuth scopes explained: a server supporting OAuth in principle isn’t the same claim as every client being able to walk through it today. If you hit an “unauthorized client” or similar error connecting Slack, that’s the most likely reason — not a mistake in your setup. Whether a given client can connect changes as vendors expand their allow-list over time, so it’s worth checking again later if it doesn’t work on your first attempt.

Option 1: Connect Slack to Claude directly

If the client you’re using is one Slack currently admits, the setup is:

  • Open Claude’s connector or MCP settings.
  • Add a remote server pointing at Slack’s MCP endpoint.
  • Sign in with your Slack account when prompted — standard OAuth, so Claude never sees your Slack password, and you can revoke access from Slack’s own admin settings at any time.

That’s enough if Claude is the only AI client you use with Slack and the connection goes through. The tradeoff shows up once you add a second client: each one needs its own attempt at the same OAuth step, with the same chance of hitting the allow-list wall independently.

Option 2: Connect it once, through a gateway

If you’re already juggling more than one MCP server or more than one AI client, wiring each pair directly gets old fast — that’s the connection sprawl covered in what an MCP gateway is. The idea is to connect Slack (and everything else) to one gateway URL, then point every AI client at that single URL instead of reconfiguring each one and re-running the OAuth step per client.

With gate specifically, connecting Slack looks like this: add gate’s gateway URL to Claude once, then pick Slack from the Slack server page in the directory and sign in the same way you would directly. gate’s catalog tracks each server’s current connectivity status — including cases like this one, where a vendor’s OAuth is still allow-listed — so you know what to expect before you try, instead of finding out mid-flow. See the full list of servers on the MCP servers page.

What to check before you connect a whole workspace

A Slack workspace usually holds more than the channel you’re thinking of when you connect it — DMs, private channels you’re a member of, and anything posted in channels the token can reach. A few things worth checking before or right after you connect:

  • Read, or read and write? A connection that can summarize a channel can also post to it — know which tools are enabled before handing Claude a prompt like “clean up the old announcements.”
  • What can it actually see? If the token covers your full workspace membership, a prompt touching one channel can surface context from others you forgot you were in.
  • Who else on the team connects it? If more than one person wires up Slack to their own client, a shared, revocable setup beats everyone holding their own grant — the same question covered in MCP access control for teams.

You can run a quick, no-signup check on any server’s tool list with gate’s free MCP security scanner, which looks for prompt injection and hidden instructions in tool descriptions before you rely on a server day to day.

Troubleshooting the connection

  • “Unauthorized client” or the OAuth screen never appears. Most likely the allow-list issue above — the client you’re using may not be one Slack currently admits for this endpoint.
  • Sign-in succeeds but Claude can’t see a channel you know exists. Slack’s tools respect your account’s own channel membership; a channel you haven’t joined won’t show up any more than it would in the Slack app itself.
  • Tools disappear after a Slack update. Remote servers can change their tool list without warning. If you’re connected through a gateway, that kind of change is exactly what drift detection is meant to catch — see how MCP rug pulls work for why a one-time review isn’t enough.
Where gate fits: gate’s directory lists each server’s currently verified connectivity — one-click OAuth, key-based, or allow-listed like Slack — so you know what kind of setup to expect before you start. Every server is scanned for prompt injection before it goes live, and every call goes into a plain-language log. See the full gateway at /mcp-gateway.

The bottom line

Connecting Slack to Claude with MCP is a normal OAuth flow when the client you’re using is one Slack currently allows — direct if it’s your only client, or once through a gateway if you want the same access shared across every AI client you use. If the connection fails at the authorization step, that’s worth reading as “not allow-listed yet” rather than a broken setup, and worth trying again as vendors open this up over time.