Convert DevOps Mattermost requests into Jira tasks with an AI sub-workflow

Overview

This is a sub-workflow that converts a free-form DevOps request posted in Mattermost into a properly formatted Jira task

Requirements

• OpenRouter/OpenAI/Anthropic API key
• Google Gemini API key — for embeddings
• Jira API credentials — Cloud or Server.
• Mattermost API credentials — to post the reply back to the channel
• Qdrant instance
• Remote MCP servers (see MCP section)
• A sub-workflow that analyses attachments
• A parent workflow that triggers this one via "Execute Workflow" with a properly shaped payload

How it works

1. The workflow is triggered by Execute Workflow Trigger
2. ReadIncidentContext logs the request and forwards the payload
3. Call 'attachmentsAnalyzer invokes a vision sub-workflow with the file_ids
4. SetVars sets workflow-level constants
5. AI Agent builds the Jira payload. The system prompt
6. Parse Agent Output safely parses the agent's output enriches the Jira description.
7. Create an issue creates a Task in the configured project with the chosen label and component.
8. Post a message replies in the original thread with a confirmation containing a clickable Jira link.

How to use

• Index your infrastructure docs into Qdrant using
• Deploy the Mattermost MCP server and put its URL into the Mattermost MCP Client node
• Deploy / import the attachmentsAnalyzer sub-workflow
• Configure credentials in the workflow
• Edit SetVars
• Edit Create an issue — pick your own Jira project, issue type and component.
• Tune the system prompt in AI Agent — adjust the direction labels and the few-shot examples to match your team's taxonomy.
• Wire this workflow into your classifier ia an Execute Workflow node