Jira Ticket Quality Review and Coaching
Monitors new Jira Stories, Bugs, and Tasks for quality, posting coaching comments only when scores fall below a configurable threshold.
What it does
Get immediate, actionable coaching comments on your Jira Stories, Bugs, and Tasks when their quality score drops below your set threshold. The system automatically analyzes new tickets using OpenRouter to provide structured feedback that helps you improve ticket clarity before they move forward.
Step by step
- Poll Jira Cloud API for new issues of type Story, Bug, or Task using a scheduled cron trigger.
- Extract summary and description fields from the retrieved JSON payload to construct the prompt input.
- Compare the LLM-generated numeric score against the configured threshold to decide whether to post a comment.
Where the LLM does the work
- Generate a nuanced coaching comment that addresses specific gaps in the ticket's acceptance criteria or clarity.
- Assign a quality score based on subjective interpretation of how well the ticket meets the configurable rubric.
- Format the output into a structured JSON object containing both the score and the feedback text.
Watch out for
Long ticket descriptions may exceed context windows, truncate or summarize input before sending to OpenRouter.
Ambiguous scoring criteria lead to inconsistent results, define explicit weighted rules in the system prompt.
Tools that fit
Jira Software Service Trigger and comment posting
OpenRouter Chat Model LLM Quality evaluation and scoring
Code API Parsing JSON response
The agent brief
Everything your agent needs, including the gotchas. Copy it and go.
You are helping me build the following AI agent workflow. ## Goal Jira Ticket Quality Review and Coaching: Monitors new Jira Stories, Bugs, and Tasks for quality, posting coaching comments only when scores fall below a configurable threshold. ## Specification - What it does: Get immediate, actionable coaching comments on your Jira Stories, Bugs, and Tasks when their quality score drops below your set threshold. The system automatically analyzes new tickets using OpenRouter to provide structured feedback that helps you improve ticket clarity before they move forward. - Trigger: Triggered by an event (Event · on new Jira issue) - Autonomy: Fully hands-off - Expected setup effort: about an afternoon - Tools/services involved: - Jira Software: Trigger and comment posting - OpenRouter Chat Model: Quality evaluation and scoring - Code: Parsing JSON response ## Known pitfalls, handle each one explicitly in your implementation 1. Long ticket descriptions may exceed context windows, truncate or summarize input before sending to OpenRouter. 2. Ambiguous scoring criteria lead to inconsistent results, define explicit weighted rules in the system prompt. ## Reference implementation https://n8n.io/workflows/16693 (workflow template) Fetch and inspect this before building. If it matches my stack, adapt it; if not, rebuild the pattern with my tools. ## Process requirements 1. Before building: ask me which of the listed tools I actually use and what my platform is (n8n / Make / code / other). Do not assume. 2. Adapt the pattern to my answers; do not force the reference stack. 3. Address every pitfall above; tell me how you handled each. 4. Provide a test plan I can run before letting this touch real data. 5. Ask before any step that sends messages, modifies data, or spends money. Source: https://usecasesforagents.com/use-case/jira-ticket-quality-review-and-coaching/ via usecasesforagents.com
Frequently asked questions
Can I use a different Project Management than Jira?
Yes. Jira is only the example project management in this recipe. The same flow works with Asana, Awork, ClickUp and Clockify. Swap the project management connection and keep the rest of the setup as written.
Want this running in your business?
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.
Who it's for
Built for developers who want the busywork around code automated, not the code itself.