Skip to content
Workflow starter brief

Chat with a self-hosted model from Slack

Routes Slack questions through Whenever to a model running on your own Ollama server and posts answers in the thread.

Build this workflowNo credit card requiredAI building has weekly usage limits.

Proposed workflow

A starting brief, not a preconfigured workflow or proof of a completed outcome. Review the generated steps and test with your accounts before publishing.

When

A mention of your bot in the Slack channel id you choose in setup.

Then

  1. Send the question to the model running on your own Ollama server.
  2. Reply in the same Slack thread with the answer, answering each mention only once.

Before you publish

Connect Slack and configure an Ollama endpoint that the hosted Whenever workflow can reach, with appropriate access controls. Inference runs on your server; messages also pass through Slack and Whenever. Self-hosting the model does not keep the conversation entirely on your network.

Connect the accounts and credentials it needs first. Optional AI steps use AI at runtime and their output can vary, so review the workflow and test it before publishing.

Runtime limits, account access and AI usage

Apps it connects

Related templates

All templates

Agents should build your automations, not be them

Describe your first workflow in plain words. One stays live for free, no credit card.

No credit card required