Sonarqube

SonarQube analysis

Analyze SonarCloud quality issues for a specific PR. Generates comprehensive report with metrics, top violators, and action plan.

Core principle: Analysis-only = no code changes, pure insight.

Process

1. Verify Token: Check $SONARQUBE_TOKEN environment variable
2. Fetch Issues: Call SonarCloud API for PR issues
3. Parse Data: Group by severity, type, file, rule
4. Generate Report: Structured output with action plan
5. Cleanup: Remove temporary files

Prerequisites

Environment variable

bashCopy code
# Set SonarQube token (add to ~/.bashrc or ~/.zshrc)
export SONARQUBE_TOKEN="your_token_here"

# Verify token is set
echo $SONARQUBE_TOKEN

To get token:

1. Go to SonarCloud → My Account → Security
2. Generate new token
3. Copy and export as environment variable

Project configuration

Configure your SonarCloud project details:

bashCopy code
# Add to project's CLAUDE.md or as environment variables
SONAR_ORGANIZATION="your-org-name"
SONAR_PROJECT_KEY="your-org_your-project"
SONAR_BASE_URL="https://sonarcloud.io/api"

If not set: Ask user to provide organization and project key.

Fetch issues

Important: Direct curl with -u "$SONARQUBE_TOKEN:" fails in zsh due to authentication parsing. Use bash script wrapper:

bashCopy code
# Create temporary bash script to handle authentication
cat > /tmp/fetch_sonar.sh << 'SCRIPT'
#!/bin/bash
curl -s -u "${SONARQUBE_TOKEN}:" \
  "https://sonarcloud.io/api/issues/search?componentKeys=${SONAR_PROJECT_KEY}&pullRequest=$1&issueStatuses=OPEN,CONFIRMED&sinceLeakPeriod=true&ps=500"
SCRIPT

chmod +x /tmp/fetch_sonar.sh
/tmp/fetch_sonar.sh $PR_NUMBER > /tmp/sonar_pr_$PR_NUMBER.json

API Parameters:

• componentKeys: Your project key
• pullRequest: PR number
• issueStatuses: OPEN,CONFIRMED (exclude resolved)
• sinceLeakPeriod: Only new issues in this PR
• ps: Page size (max 500)

Analysis script

Create Node.js analysis script at /tmp/sonar_analyze.js:

javascriptCopy code
const fs = require('fs');
const prNumber = process.argv[2];
const data = JSON.parse(fs.readFileSync(`/tmp/sonar_pr_${prNumber}.json`, 'utf8'));
const issues = data.issues || [];

// Group by severity
const bySeverity = issues.reduce((acc, i) => {
  acc[i.severity] = (acc[i.severity] || 0) + 1;
  return acc;
}, {});

// Group by type
const byType = issues.reduce((acc, i) => {
  acc[i.type] = (acc[i.type] || 0) + 1;
  return acc;
}, {});

// Group by file
const byFile = issues.reduce((acc, i) => {
  const file = i.component.split(':')[1] || i.component;
  acc[file] = (acc[file] || 0) + 1;
  return acc;
}, {});

// Group by rule
const byRule = issues.reduce((acc, i) => {
  if (!acc[i.rule]) {
    acc[i.rule] = {
      count: 0,
      severity: i.severity,
      message: i.message
    };
  }
  acc[i.rule].count++;
  return acc;
}, {});

// Output structured data
console.log(JSON.stringify({
  total: data.total,
  bySeverity,
  byType,
  topFiles: Object.entries(byFile)
    .sort((a, b) => b[1] - a[1])
    .slice(0, 10),
  topRules: Object.entries(byRule)
    .map(([rule, d]) => ({ rule, ...d }))
    .sort((a, b) => b.count - a.count)
    .slice(0, 5)
}, null, 2));

Run analysis:

bashCopy code
node /tmp/sonar_analyze.js $PR_NUMBER > /tmp/sonar_analysis_$PR_NUMBER.json

Report format

Generate formatted report from analysis:

Copy code
📊 SonarCloud Analysis - PR #XXX

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
📈 EXECUTIVE SUMMARY
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Total Issues: {TOTAL}

By Severity:
🔴 Blocker/Critical: {COUNT} ({PERCENTAGE}%)
🟡 Major: {COUNT} ({PERCENTAGE}%)
🔵 Minor/Info: {COUNT} ({PERCENTAGE}%)

By Type:
🐛 Bugs: {COUNT}
🛡️ Vulnerabilities: {COUNT}
🧹 Code Smells: {COUNT}

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
📂 TOP 10 FILES WITH ISSUES
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

1. src/components/UserProfile.tsx - 8 issues
2. src/services/auth.service.ts - 5 issues
3. src/utils/validation.ts - 4 issues
...

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
⚠️ TOP 5 VIOLATED RULES
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

1. typescript:S1854 (MAJOR) - 12 occurrences
   "Dead stores should be removed"

2. typescript:S3776 (CRITICAL) - 8 occurrences
   "Cognitive Complexity of functions should not be too high"

3. typescript:S1186 (MINOR) - 6 occurrences
   "Functions should not be empty"
...

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
✅ ACTION PLAN
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Priority 1 - CRITICAL/BLOCKER ({COUNT} issues):
  • Fix immediately before merge
  • Focus on: {TOP_FILES}

Priority 2 - MAJOR ({COUNT} issues):
  • Address in this PR if possible
  • Consider technical debt ticket if extensive

Priority 3 - MINOR/INFO ({COUNT} issues):
  • Can be addressed in follow-up PR
  • Add to backlog for refactoring sprint

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
🔗 LINKS
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

View in SonarCloud:
https://sonarcloud.io/project/pull_requests_list?id={PROJECT_KEY}&pullRequest={PR_NUMBER}

Severity mapping

Issue types

Cleanup

Always clean up temporary files after execution:

bashCopy code
rm -f /tmp/fetch_sonar.sh
rm -f /tmp/sonar_pr_$PR_NUMBER.json
rm -f /tmp/sonar_analyze.js
rm -f /tmp/sonar_analysis_$PR_NUMBER.json

Error handling

Configuration options

Project-level configuration

Create .sonarcloud.properties or add to CLAUDE.md:

propertiesCopy code
# SonarCloud Configuration
SONAR_ORGANIZATION=your-org
SONAR_PROJECT_KEY=your-org_your-project
SONAR_EXCLUSIONS=**/*.test.ts,**/*.spec.ts,**/migrations/**
SONAR_COVERAGE_EXCLUSIONS=**/*.test.ts,src/test/**

API rate limits

SonarCloud API limits:

• Free tier: 10,000 requests/day
• Paid tier: Unlimited

Tip: Cache results for repeated queries to same PR.

Integration examples

GitHub Actions

yamlCopy code
- name: SonarQube Analysis
  run: |
    export SONARQUBE_TOKEN=${{ secrets.SONAR_TOKEN }}
    export SONAR_PROJECT_KEY="${{ secrets.SONAR_PROJECT }}"
    claude -p "/sonarqube ${{ github.event.pull_request.number }}"

Pre-merge Hook

Add to .claude/hooks/pre-merge.sh:

bashCopy code
#!/bin/bash
PR_NUMBER=$(gh pr view --json number -q .number)
claude -p "/sonarqube $PR_NUMBER"

Red flags - NEVER do

Never:

• ❌ Modify code or auto-fix issues (analysis-only command)
• ❌ Skip token verification (security risk)
• ❌ Leave temp files in /tmp (cleanup required)
• ❌ Commit SonarQube token to repository (use env vars)
• ❌ Run without checking token expiration

Always:

• ✅ Generate structured, actionable report
• ✅ Clean up after execution
• ✅ Handle API errors gracefully
• ✅ Verify token is valid before API calls
• ✅ Parse and present data clearly

Advanced usage

Custom filters

bashCopy code
# Only show critical/blocker issues
/sonarqube 123 --severity BLOCKER,CRITICAL

# Only show bugs and vulnerabilities
/sonarqube 123 --types BUG,VULNERABILITY

# Specific file pattern
/sonarqube 123 --files "src/services/**"

Multiple PRs

bashCopy code
# Compare issues across PRs
/sonarqube 123,124,125 --compare

Troubleshooting

Issue: "curl: (22) The requested URL returned error: 401"

Cause: Invalid or missing token

Fix:

bashCopy code
# Regenerate token in SonarCloud
# Export new token
export SONARQUBE_TOKEN="new_token_here"

Issue: "Empty response or no issues"

Cause: Analysis not yet complete or PR not analyzed

Fix: Wait for SonarCloud analysis to complete (~2-5 minutes after PR creation)

Issue: "componentKeys not found"

Cause: Wrong project key

Fix: Verify project key in SonarCloud URL:

Copy code
https://sonarcloud.io/project/overview?id=YOUR_PROJECT_KEY

Usage examples

bashCopy code
# Basic usage
/sonarqube 170

# With PR prefix
/sonarqube PR #234

# Using PR URL
/sonarqube https://github.com/org/repo/pull/170

# Custom severity filter (if implemented)
/sonarqube 170 --critical-only

PR Number: $ARGUMENTS