defect-lifecycle-manager（缺陷生命周期经理）

Defect Lifecycle Manager

Expert skill for managing defects throughout their lifecycle and tracking quality metrics.

When to Use

Use this skill when you need to:

• Log new defects with complete information
• Track defect status and lifecycle
• Analyze defect trends and metrics
• Conduct root cause analysis
• Report quality status to stakeholders
• Verify defect fixes
• Prevent defect recurrence

Defect Lifecycle States

NEW → ASSIGNED → IN PROGRESS → RESOLVED → VERIFIED → CLOSED
                                    ↓
                                REOPENED (if verification fails)

State Definitions

NEW: Just logged, awaiting triage ASSIGNED: Assigned to developer IN PROGRESS: Developer actively working RESOLVED: Fix completed, ready for verification VERIFIED: QA confirmed fix works CLOSED: Accepted and archived REOPENED: Fix didn't work, back to dev

Defect Logging Standards

Complete Defect Report

Defect ID: [Auto-generated or manual]
Title: [Action] + [Component] + [Unexpected Behavior]
  Example: "Login button does not respond on mobile Chrome"

Type: [Functional/UI/Performance/Security/Data/Integration]
Severity: [Critical/High/Medium/Low]
Priority: [P0/P1/P2/P3]

Reported By: [Name]
Date Reported: [Date]
Detected In: [Environment]
Build/Version: [x.y.z]

Environment Details:
- OS: [Platform + version]
- Browser/Device: [Details]
- Network: [WiFi/Mobile/etc]

Preconditions:
[System state before reproduction]

Steps to Reproduce:
1. [Exact action]
2. [Exact action]
3. [Exact action]

Expected Result:
[What should happen per requirements]

Actual Result:
[What actually happened]

Reproducibility: [Always / Intermittent (X out of Y attempts)]

Impact:
[Who is affected? How many users? What's blocked?]

Attachments:
- Screenshot: [filename]
- Error log: [filename]
- Video: [filename]
- Network trace: [filename]

Additional Context:
- Related defects: [Links]
- Related requirements: [Links]
- Workaround: [If available]

Defect Triage Process

Step 1: Validate

• Can you reproduce it?
• Is it a duplicate?
• Is it really a defect or expected behavior?

Step 2: Categorize

• Determine type and root cause area
• Assign to correct component/team

Step 3: Assess Impact

• Severity (technical impact)
• Priority (business impact + timing)
• Customer impact

Step 4: Assign & Schedule

• Route to appropriate developer
• Set target fix version
• Establish timeline

Defect Metrics & Analysis

Key Metrics

Defect Density

Defects per 1000 lines of code
or
Defects per feature/module

Defect Detection Rate

(Defects found in QA) / (Total defects found)
Goal: 90%+ caught before production

Defect Age

Days from New → Closed
Track: Average, Median, Max

Fix Time

Days from Assigned → Resolved
By severity level

Reopen Rate

(Reopened defects) / (Total resolved)
Goal: < 10%

Defect Escape Rate

(Production defects) / (Total defects)
Goal: < 5%

Trend Analysis

Monitor weekly/monthly:

• Defect discovery rate (increasing/decreasing?)
• Open vs closed trend (gap closing or widening?)
• Severity distribution (more critical over time?)
• Component hot spots (which areas have most defects?)
• Root cause patterns (similar issues repeating?)

Root Cause Analysis

5 Whys Technique

Problem: Users unable to login

Why? → Authentication service returned error 500
Why? → Database connection timed out
Why? → Connection pool exhausted
Why? → Pool size set too low for expected load
Why? → Configuration not updated after scaling users

Root Cause: Incorrect configuration management
Prevention: Add load testing + auto-scaling configs

Common Root Causes

1. Requirements Issues

- Missing information - Ambiguous specifications - Changing requirements

1. Design Issues

- Incorrect architecture - Missing edge case handling - Performance not considered

1. Implementation Issues

- Coding errors - Logic mistakes - Missing validation

1. Testing Gaps

- Test cases missing scenarios - Insufficient coverage - Environment differences

1. Environmental Issues

- Configuration problems - Deployment errors - Integration failures

Defect Prevention

Based on root cause analysis:

• Update Requirements: Add missing acceptance criteria
• Enhance Test Cases: Cover discovered edge cases
• Improve Code Reviews: Add checklists for common issues
• Automate Detection: Create automated tests
• Team Training: Share lessons learned
• Process Improvements: Update development practices

Defect Verification

When defect is marked RESOLVED:

Verification Steps

1. Review Fix: Understand what was changed
2. Verify Original Issue: Run original reproduction steps
3. Test Related Areas: Check for unintended side effects
4. Regression Test: Ensure fix doesn't break other features
5. Edge Cases: Test variations of original scenario
6. Document Result: Update defect with verification notes

Verification Outcomes

• ✓ VERIFIED: Fix works, no side effects → CLOSED
• ✗ REOPENED: Issue still occurs or new issues introduced
• ⚠️ PARTIAL: Fixed but with new concerns → Discuss with dev

Defect Status Reports

Daily Standup Update

Yesterday: Fixed 5 defects, found 3 new
Today: Will verify 8 fixes, test new feature
Blockers: 2 critical defects blocking release

Weekly Status Report

Week: [Date range]

Defect Summary:
- Opened: 45
- Closed: 38
- Net Change: +7

Open Defects: 67 total
- Critical: 2
- High: 12
- Medium: 35
- Low: 18

Top Risks:
1. [Critical defect description + impact]
2. [High priority issue + timeline]

Trends:
- Defect rate decreasing (good)
- Backlog growing (concerning)
- Most defects in payment module (focus area)

Release Readiness Report

Release: v2.3.0
Date: [Target date]

Status: [GREEN / YELLOW / RED]

Defect Gate Criteria:
✓ Critical defects: 0 (Goal: 0)
✓ High defects: 2 (Goal: ≤3)
✗ P0/P1 open: 1 (Goal: 0) ← BLOCKER

Recommendation: [Go / No-Go / Conditional Go]

Conditions for Go:
1. [Must fix BUG-XXX by date]
2. [Must complete regression testing]

Best Practices

• ✓ Log defects immediately, don't batch
• ✓ Be specific and factual, not subjective
• ✓ Include all reproduction steps
• ✓ Attach evidence (screenshots, logs)
• ✓ Follow up on assigned defects
• ✓ Verify fixes thoroughly
• ✓ Close defects promptly
• ✓ Analyze patterns for prevention
• ✓ Communicate risks early
• ✓ Keep stakeholders informed