Every 10 minutes, the agent finds starred Slack messages, generates a title and tags via AI, and saves them to your Notion database.
Verified 2026-07-07
Ops-ITDeveloper-Engineering
What it does
Capture important Slack messages instantly by having them organized into your Notion database every ten minutes. You get AI-generated titles and relevant tags for each starred message without any manual sorting.
How it works
When it runs
On a schedule · Scheduled · every 10 minutes
How hands-off
Assists you, you do the final step
Setup
An afternoon · about an afternoon
Works with
Slack, Notion, OpenAI Chat Model
Tools that fit
Services
Slack : Fetching messages and detecting star reactions
Notion : Storing archived messages and tags in a database
LLM
OpenAI Chat Model : Generating titles and categorizing messages with AI
Watch out for
⚠ Slack API rate limits may block bulk processing, so add exponential backoff logic to handle throttling gracefully.
⚠ Notion property types must match the data format exactly, so ensure string tags are converted to multi-select options before creation.
⚠ LLM outputs can vary in structure, so implement robust error handling and fallback defaults if the model returns malformed JSON.
⚠ Duplicate messages might be processed repeatedly, so add a check against existing Notion pages using the Slack message ID as a unique key.
This is what I do. I design and build AI agents like this one, and keep them running. If you want it set up for your team instead of doing it yourself, get in touch.