[BUG] Slack MCP "Failed to connect" — cannot reconnect after re-authentication
Preflight Checklist
- [x] I have searched existing issues and this hasn't been reported yet
- [x] This is a single bug report (please file separate reports for different bugs)
- [x] I am using the latest version of Claude Code
What's Wrong?
▎ Issue: Slack MCP "Failed to connect" — cannot reconnect after re-authentication
▎ Error: {"jsonrpc":"2.0","id":null,"error":{"code":-32001,"message":"missing_token"}}
▎ Reference ID: ofid_927339d69f9098ff
▎ Details:
▎ - Other MCP servers (Clockwise, Notion, Gmail, Google Calendar, Miro) all connect successfully
▎ - Slack only returns missing_token after OAuth flow
▎ - Tried: disable/enable, remove/re-add from Claude Desktop, re-authenticate multiple times
▎ - Endpoint https://mcp.slack.com/mcp returns HTTP 401 (reachable)
What Should Happen?
repair
Error Messages/Logs
{"jsonrpc":"2.0","id":null,"error":{"code":-32001,"message":"missing_token"}}
Steps to Reproduce
▎ Details:
▎ - Other MCP servers (Clockwise, Notion, Gmail, Google Calendar, Miro) all connect successfully
▎ - Slack only returns missing_token after OAuth flow
▎ - Tried: disable/enable, remove/re-add from Claude Desktop, re-authenticate multiple times
▎ - Endpoint https://mcp.slack.com/mcp returns HTTP 401 (reachable)
Claude Model
None
Is this a regression?
No, this never worked
Last Working Version
_No response_
Claude Code Version
latest
Platform
Anthropic API
Operating System
macOS
Terminal/Shell
Terminal.app (macOS)
Additional Information
_No response_
8 Comments
Found 3 possible duplicate issues:
This issue will be automatically closed as a duplicate in 3 days.
🤖 Generated with Claude Code
Same issue here — additional findings
Experiencing the exact same problem. Slack MCP is the only one failing while Notion, Linear, Gmail, and Google Calendar MCP all work fine.
Key finding: v2.1.80 → v2.1.81 regression
Slack MCP worked on v2.1.80, but broke after upgrading to v2.1.81. This strongly suggests a regression introduced in v2.1.81.
The most likely culprit is the CIMD (SEP-991) support added in v2.1.81 for MCP OAuth with servers that don't support Dynamic Client Registration — which includes Slack's MCP server (
mcp.slack.com).My symptoms
| Item | Status |
|------|--------|
| Auth | ✔ authenticated |
| Status | ✘ failed |
| Error |
App is not enabled for Slack MCP server access|| Error URL |
https://api.slack.com/apps/A08SF47R6P4/app-assistant→ 404 || Claude chat in Slack | ✔ works fine |
| Notion/Linear/Gmail MCP | ✔ all working |
What I've tried (all ineffective)
clientId: 1601185624273.8899143856786)Environment
Analysis
The error
"App is not enabled for Slack MCP server access"points to the app-level MCP access gate (post-OAuth), not the OAuth authentication itself. The v2.1.81 CIMD/SEP-991 change likely altered how the MCP client presents its identity tomcp.slack.com, causing Slack's server to reject the connection at the MCP enablement check.Can confirm that I'm seeing the same thing and came to the same result as @nishijun - my logs and output in Claude code confirm the authentication is successful, but the server reconnection fails with that same "App is not enabled for access" message. I've also followed the same "remove it and reconnect it" steps to the same ineffective result.
Environment
Same issue as well. It says here "Authorization with the MCP server failed. You can check your credentials and permissions."
Environment
Same issue — OAuth redirect_uri mismatch
Authentication fails with:
Slack MCP shows as "Connected" but no tools are exposed. Other MCP servers (Figma, Atlassian, Firebase) all work fine.
Environment
Same issue -- seems to affect other similarly situated mcp tools as well.
Same here
<img width="1576" height="988" alt="Image" src="https://github.com/user-attachments/assets/c08093fb-2150-472b-9ab7-e62d576f3c59" />
Update: Slack MCP is working again 🎉
As of ~2026-03-24 06:50 UTC+8, Slack MCP is back and functioning normally. Search tools are returning results successfully.
No changes were made on my end — this appears to have been fixed server-side by Anthropic/Slack.
Environment (unchanged):
Can others confirm?