Slack connector stopped working after an hour? Reconnect in Sume
When Slack returns no token expiry, Sume assumes one hour. After that an admin must connect again, because v1 has no automatic token renewal for integrations.
If the Slack tools in your Sume workspace stop working, the token has probably expired. Sume uses a one-hour fallback when Slack gives no provider expiry, and it does not renew tokens automatically. An admin must open Integrations and connect Slack again. Reconnecting starts a fresh consent and replaces the stored token. No other setting needs to change.
What the docs say
The workspace integrations architecture doc states that tokens without a provider expiry use the one-hour fallback, that after that time an admin must connect again, and that Sume does not implement automatic token renewal. The same rule appears for tool execution generally: each call reads the workspace connection again and checks token expiry, and in v1 an expired token means connect again.
What you will notice
- Slack tools that worked earlier fail or disappear in a new agent turn
- The Integrations page is the place to check the connection and reconnect
- A plugin outage does not break unrelated agent tools; discovery for other tools keeps working
How to reconnect
Sign in as an admin of the workspace, open /dashboard/integrations, and use the Slack row to start consent again. The callback checks admin membership for the original workspace, not the Clerk organization currently selected, so start the flow in the workspace you want to fix.
Is this the same as MCP OAuth?
No. This is the connector token Sume holds for Slack. The hosted MCP login you use from Claude or Cursor is a different token with its own one-hour life; see Sume's one-hour MCP token. If you want a credential that does not expire in the middle of a session, use an API key for the MCP client. That choice does not change how the Slack connector behaves.
Plan for it
Treat a Slack-reading agent as something that needs a human to reconnect now and then. Do not build an unattended nightly job that depends on the Slack tools staying alive for days.
A short runbook
When someone reports that the Slack tools are gone, first check the Integrations page for the Slack row. If it looks connected but calls fail, treat the token as expired and reconnect as an admin. Then start a new agent turn in the same workspace, because the tool list is built per turn.
Finally, note the time you reconnected. With a one-hour fallback and no renewal, the next reconnect is due about an hour later if Slack gives no expiry, so schedule human attention rather than a background job.
Related posts
More in Integrations
- Slack MCP has no dynamic registration: how Sume connects anyway
Slack's MCP server needs a registered Slack app with a client secret and does not support Dynamic Client Registration, so Sume uses its own Slack app.
- Slack consent worked but MCP rejects the app: enable Slack MCP
If Slack OAuth consent succeeds but MCP initialization is rejected, the Slack app has not enabled the Slack MCP server. It is a separate switch from Agents.
- Slack slash command for music: 3 s ack, 5 replies in 30 min
A Slack command must be acknowledged in 3 s; response_url then allows 5 replies in 30 min. Start the Sume music job in webhook mode and reply when it lands.
- Slack Workflow webhook trigger: flatten a Sume result, no nested JSON
Slack's webhook-started workflows take up to 20 flat variables and 1 request per second. Flatten Sume's nested artifacts into plain text values before you post.
Written by Sume