Jira Triage Automation with OpenAI, Notion & Slack (n8n)

Automatically triage stale Jira tickets with OpenAI, Notion, and Slack (n8n)

This scheduled n8n workflow scans your Jira issues that have lingered, uses OpenAI to classify status and sentiment, searches Notion (and similar Jira issues) for answers, and then posts reminders or resolutions. When a human needs to step in, it alerts your Slack channel—so support and product teams waste less time on manual follow-up.

What this workflow does

• Runs on a schedule and queries Jira for issues in “To Do” or “In Progress” created at least 7 days ago.
• Processes each ticket individually by collecting issue metadata and all Jira comments, then formatting the thread into AI-friendly text.
• Uses OpenAI classification to determine whether the ticket is resolved, pending more information, or still waiting on a response.
• For resolved tickets: runs OpenAI sentiment analysis to decide whether to request feedback and close, autoclose with a neutral message, or alert Slack when sentiment is negative.
• For tickets still waiting: checks that the last comment is not an automated message, then uses OpenAI to generate and post a Jira reminder comment.
• For pending-more-information tickets: uses an OpenAI agent to search Notion and similar Jira issues, then either posts the found answer and closes the ticket, or notifies Slack and still closes.

Use cases

• Keep Jira hygiene by automatically escalating long-lived “To Do” and “In Progress” items.
• Reduce support turnaround by reminding stakeholders and posting AI-generated follow-ups.
• Improve resolution consistency by retrieving prior answers from Notion and existing Jira threads.
• Surface negative customer/service sentiment to a Slack channel for manual attention.

Technical details

• Integrations: Jira, OpenAI, Notion, Slack.
• n8n nodes mentioned: if, set, jira, jira tool, slack, aggregate.
• Setup requirements: Jira credentials with permission to read issues/comments, add comments, and transition issues to your “Done” status; plus an OpenAI API credential for the classification models used.