Jira Ticket Analysis Skill

This skill provides methodology for analyzing Jira ticket quality and identifying areas for improvement.

Quality Dimensions

1. Clarity

Questions to assess:

• Is the goal obvious from the summary?
• Is the description free of jargon?
• Would a new team member understand this?
• Are acronyms explained?

Scoring:

• 5: Crystal clear, no questions needed
• 4: Clear with minor ambiguity
• 3: Understandable but needs context
• 2: Confusing, multiple interpretations
• 1: Cannot understand the request

2. Completeness

Questions to assess:

• Are acceptance criteria defined?
• Is the scope clear (what's in/out)?
• Are dependencies listed?
• Is technical context provided?

Scoring:

• 5: All information present, ready to implement
• 4: Minor details missing
• 3: Key information missing but discoverable
• 2: Significant gaps
• 1: Missing most required information

3. Actionability

Questions to assess:

• Can work start immediately?
• Are blockers identified and addressed?
• Is the scope achievable?
• Are success criteria measurable?

Scoring:

• 5: Can start now with confidence
• 4: One small clarification needed
• 3: Some research/clarification needed
• 2: Blocked or unclear how to proceed
• 1: Cannot start without major discovery

4. Context

Questions to assess:

• Is the "why" explained?
• Is user impact described?
• Are related tickets linked?
• Is there technical background?

Scoring:

• 5: Full context, understand importance
• 4: Good context, minor gaps
• 3: Basic context provided
• 2: Little context, unclear importance
• 1: No context, don't know why this matters

Analysis Process

Step 1: Initial Read

• Read summary and description
• Note first impressions
• Identify obvious issues

Step 2: Score Dimensions

Rate each dimension 1-5 and note specific issues:

| Dimension | Score | Issues |
|-----------|-------|--------|
| Clarity | 3 | Acronyms undefined, vague scope |
| Completeness | 2 | No acceptance criteria |
| Actionability | 2 | Blocked by design |
| Context | 4 | Good background |
| **Overall** | **2.75** | Needs refinement |

Step 3: Identify Improvements

For each issue, provide:

• What's wrong
• How to fix it
• Example improved text

Step 4: Provide Recommendations

Prioritize fixes by impact:

1. Critical (blocks understanding)
2. Important (delays work)
3. Nice-to-have (polish)

Common Issues and Fixes

Issue: Vague Summary

Before: "Fix the thing" After: "Fix payment processing timeout for orders over $1000"

Pattern: [Action] [specific thing] [condition/context]

Issue: Missing Acceptance Criteria

Before:

Make the page faster

After:

h2. Acceptance Criteria
* Page load time < 2 seconds on 3G connection
* Lighthouse performance score > 80
* No layout shift during load
* Images lazy-loaded below fold

Issue: No Context/Why

Before:

Add retry logic to API calls

After:

h2. Context
Users are experiencing intermittent failures when the payment service is slow to respond. This causes checkout abandonment and support tickets.

h2. Problem
When the payment API times out (typically 2-3 times per day during peak hours), the user sees an error with no recovery path.

h2. Solution
Add retry logic with exponential backoff to handle transient failures gracefully.

Issue: Code-Heavy Description

Before:

Change this:
function processOrder(order) {
  // 50 lines of code
}

To this:
function processOrder(order) {
  // 50 different lines of code
}

After:

h2. Change Required
Modify {{src/services/orderProcessor.ts}} to validate order totals before processing.

h2. Technical Notes
* Add validation in processOrder function
* Throw OrderValidationError for invalid totals
* Log validation failures for monitoring

Issue: Scope Creep

Before:

Add user notifications. Also we should probably redo the entire notification system and add email preferences and maybe webhooks too.

After:

h2. Scope
Add email notification when order ships.

h2. Out of Scope (Future Tickets)
* Notification preferences UI
* Webhook integrations
* Push notifications

Analysis Report Template

## Ticket Quality Analysis: PROJ-123

### Summary
[Brief description of the ticket]

### Quality Scores

| Dimension | Score | Status |
|-----------|-------|--------|
| Clarity | X/5 | [Good/Needs Work/Poor] |
| Completeness | X/5 | [Good/Needs Work/Poor] |
| Actionability | X/5 | [Good/Needs Work/Poor] |
| Context | X/5 | [Good/Needs Work/Poor] |
| **Overall** | **X/5** | |

### Issues Found

#### Critical Issues
1. [Issue description]
   - Impact: [Why this matters]
   - Fix: [How to resolve]

#### Important Issues
1. [Issue description]
   - Impact: [Why this matters]
   - Fix: [How to resolve]

#### Minor Issues
1. [Issue description]

### Recommended Improvements

1. **[Most Important Fix]**
   Current: [What it says now]
   Improved: [What it should say]

2. **[Second Fix]**
   [Details]

### Readiness Assessment
- [ ] Ready for sprint planning
- [x] Needs refinement first
- [ ] Needs product clarification
- [ ] Needs technical discovery

Batch Analysis Patterns

Finding Low-Quality Tickets

# Missing description
project = PROJ AND description IS EMPTY

# Short descriptions (< 100 chars approximation)
project = PROJ AND description ~ "[a-zA-Z]" AND NOT description ~ "*\\n*"

# No acceptance criteria pattern
project = PROJ AND description !~ "acceptance" AND description !~ "criteria" AND type = Story

# Old unrefined tickets
project = PROJ AND status = "To Do" AND created <= -60d AND labels NOT IN ("refined", "ready")

Quality Dashboard Metrics

Track over time:

• % tickets with acceptance criteria
• Average quality score
• Tickets needing refinement
• Time from creation to ready status

Integration with Refinement

Pre-Refinement Analysis

1. Run batch analysis on sprint candidates
2. Generate quality report
3. Prioritize tickets needing work
4. Estimate refinement effort

During Refinement

1. Review analysis findings
2. Discuss improvements with team
3. Update tickets in real-time
4. Re-score after improvements

Post-Refinement Verification

1. Verify all critical issues addressed
2. Confirm acceptance criteria added
3. Check technical context included
4. Mark tickets as "refined"