Give your IT team an AI helper inside Slack. When someone sends a direct message, it replies fast, looks up answers in your knowledge base, and keeps the chat clean. It is built for helpdesks that want quicker replies without extra manual work.

Incoming DMs hit an n8n Webhook, and a Verify Webhook node answers Slack’s challenge so the app stays active. Messages from bots are skipped. The user gets a short receipt message first. An AI Agent then builds the final answer using the OpenAI Chat Model and a memory window that keeps recent context per channel. The agent can call a tool that connects to a second n8n workflow which searches Confluence by keywords. The message is cleaned to match Slack formatting, the initial receipt is deleted, and the final reply is posted to the same DM.

Setup needs a Slack app with Events API, an OpenAI API key, and a Confluence space if you want knowledge lookups. Expect faster first responses, less context switching, and fewer repeated questions. This works well for IT help, onboarding questions, and policy lookups, all inside Slack.

• Slack Events webhook receives DMs and a verify node answers Slack’s challenge
• Bot filter skips messages sent by other apps to save processing
• Instant acknowledgment message with a custom bot icon to reassure users
• AI Agent powered by an OpenAI Chat Model that you can change when needed
• Memory buffer keeps the last 10 messages per channel for context
• Tool call to a second workflow that searches Confluence by keywords
• Output cleanup converts Markdown links and bold text to proper Slack format
• Deletes the initial receipt and posts the final answer in the same DM

• Cut first response time from minutes to seconds
• Automate common IT questions using your knowledge base
• Keep direct messages tidy by removing receipt messages
• Support more requests at once without extra staff
• Connect Slack, OpenAI, and Confluence in one flow
• Maintain conversation context for more accurate answers

1. Import the template into n8n: Create a new workflow in n8n > Click the three dots menu > Select 'Import from File' > Choose the downloaded JSON file.
2. You'll need accounts with Slack, OpenAI and Confluence. See the Tools Required section above for links to create accounts with these services.
3. In n8n, open any Slack node (Send Initial Message or Send Message). In the 'Credential to connect with' menu, click 'Create new credential', then follow the on screen steps to connect your Slack app.
4. Open the OpenAI Chat Model node. In the 'Credential to connect with' dropdown, click 'Create new credential'. Generate an API key in your OpenAI account and paste it into n8n.
5. In your Slack app settings, enable Events API. Set the Request URL to the 'Receive DMs' webhook URL shown in n8n and wait for Slack to verify it. Subscribe to message.im events and install the app to your workspace.
6. If you use production and test URLs in n8n, copy the Production URL from the 'Receive DMs' node and update the Slack Request URL after testing.
7. Open the AI Agent node and confirm the model is set. You can switch models by editing the connected Chat Model node.
8. To enable knowledge lookups, create a second n8n workflow that calls the Confluence search API with an HTTP Request node. Copy its workflow ID and paste it into the 'Call Confluence Workflow Tool' node field.
9. Send a test DM to your Slack app. You should see a short acknowledgment first and then the final AI answer. Check n8n Executions for logs if nothing arrives.
10. Troubleshooting: If Slack cannot verify the webhook, make sure your n8n URL is public and the Verify Webhook node returns the challenge. If messages fail to send, check Slack permissions and confirm the channel ID in each Slack node is mapped from the incoming event.