feat: add visor-agent-dev skill for AI coding assistants (#562) #2119
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: 🤖 AI Battle Test | ||
|
Check failure on line 1 in .github/workflows/ai-battle-test.yml
|
||
| on: | ||
| pull_request: | ||
| types: [opened, synchronize, edited] | ||
| issue_comment: | ||
| types: [created] | ||
| workflow_dispatch: | ||
| inputs: | ||
| test_scenario: | ||
| description: 'Test scenario to run' | ||
| required: false | ||
| default: 'all' | ||
| type: choice | ||
| options: | ||
| - 'all' | ||
| - 'incremental_analysis' | ||
| - 'xml_formatting' | ||
| - 'comment_management' | ||
| - 'provider_types' | ||
| # Cancel previous in-progress runs of this workflow on the same ref (branch/PR) | ||
| concurrency: | ||
| group: ai-battle-${{ github.workflow }}-${{ github.ref }} | ||
| cancel-in-progress: true | ||
| permissions: | ||
| contents: read | ||
| pull-requests: write | ||
| issues: write | ||
| checks: write | ||
| jobs: | ||
| ai-battle-test: | ||
| name: AI Battle Test — ${{ matrix.ai_provider.name }} — ${{ github.event_name }} | ||
| runs-on: ubuntu-latest | ||
| if: > | ||
| github.event_name == 'pull_request' || | ||
| github.event_name == 'workflow_dispatch' || | ||
| ( | ||
| github.event_name == 'issue_comment' && | ||
| github.event.issue.pull_request && | ||
| contains(github.event.comment.body, '/ai-test') | ||
| ) | ||
| strategy: | ||
| fail-fast: false | ||
| matrix: | ||
| ai_provider: | ||
| - name: 'Google Gemini' | ||
| api_key_secret: 'GOOGLE_API_KEY' | ||
| model: 'gemini-2.0-flash-exp' | ||
| env_var: 'GOOGLE_API_KEY' | ||
| - name: 'OpenAI GPT-4' | ||
| api_key_secret: 'OPENAI_API_KEY' | ||
| model: 'gpt-4o-mini' | ||
| env_var: 'OPENAI_API_KEY' | ||
| - name: 'Anthropic Claude' | ||
| api_key_secret: 'ANTHROPIC_API_KEY' | ||
| model: 'claude-3-haiku-20240307' | ||
| env_var: 'ANTHROPIC_API_KEY' | ||
| - name: 'Mock AI (Fallback)' | ||
| api_key_secret: '' | ||
| model: 'mock' | ||
| env_var: '' | ||
| steps: | ||
| - name: 🛒 Checkout code | ||
| uses: actions/checkout@v4 | ||
| with: | ||
| fetch-depth: 0 | ||
| # For PR events, checkout the PR head | ||
| ref: ${{ github.event_name == 'pull_request' && github.head_ref || github.ref }} | ||
| - name: 🔧 Setup Node.js | ||
| uses: actions/setup-node@v4 | ||
| with: | ||
| node-version: '20' | ||
| cache: 'npm' | ||
| - name: 📦 Install dependencies | ||
| run: npm ci | ||
| - name: 🏗️ Build Visor | ||
| run: npm run build | ||
| - name: 🧪 Test 1 - XML Formatted PR Analysis | ||
| id: xml_test | ||
| uses: ./ | ||
| with: | ||
| github-token: ${{ secrets.GITHUB_TOKEN }} | ||
| auto-review: true | ||
| add-reactions: 'false' | ||
| env: | ||
| # Multiple API keys for fallback - similar to probe.yml approach | ||
| GOOGLE_API_KEY: ${{ secrets.GOOGLE_API_KEY }} | ||
| GEMINI_API_KEY: ${{ secrets.GOOGLE_API_KEY }} # Aider expects GEMINI_API_KEY | ||
| OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }} | ||
| ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }} | ||
| # Custom API URLs for quota management | ||
| GOOGLE_API_URL: ${{ secrets.GOOGLE_API_URL }} | ||
| ANTHROPIC_API_URL: ${{ secrets.ANTHROPIC_API_URL }} | ||
| ANTHROPIC_BASE_URL: ${{ secrets.ANTHROPIC_API_URL }} | ||
| OPENAI_API_URL: ${{ secrets.OPENAI_API_URL }} | ||
| LLM_BASE_URL: ${{ secrets.LLM_BASE_URL }} | ||
| # Model and provider configuration | ||
| MODEL_NAME: ${{ matrix.ai_provider.model }} | ||
| FORCE_PROVIDER: ${{ secrets.FORCE_PROVIDER }} | ||
| # GitHub context | ||
| GITHUB_CONTEXT: ${{ toJson(github) }} | ||
| continue-on-error: true | ||
| - name: 📊 Validate XML Test Results | ||
| if: steps.xml_test.outcome == 'success' | ||
| run: | | ||
| echo "✅ XML Formatted Analysis completed successfully" | ||
| echo "Provider: ${{ matrix.ai_provider.name }}" | ||
| echo "Total Issues: ${{ steps.xml_test.outputs.total-issues }}" | ||
| echo "Critical Issues: ${{ steps.xml_test.outputs.critical-issues }}" | ||
| echo "PR Action: ${{ steps.xml_test.outputs.pr-action }}" | ||
| echo "Incremental Analysis: ${{ steps.xml_test.outputs.incremental-analysis }}" | ||
| # Validate presence of total-issues for non-mock providers | ||
| if [ -z "${{ steps.xml_test.outputs.total-issues }}" ] && [ "${{ matrix.ai_provider.name }}" != "Mock AI (Fallback)" ]; then | ||
| echo "❌ Missing total-issues output" | ||
| exit 1 | ||
| fi | ||
| - name: 🧪 Test 2 - Incremental Analysis (Simulate New Commit) | ||
| if: github.event_name == 'pull_request' && github.event.action == 'synchronize' | ||
| id: incremental_test | ||
| uses: ./ | ||
| with: | ||
| github-token: ${{ secrets.GITHUB_TOKEN }} | ||
| auto-review: true | ||
| add-reactions: 'false' | ||
| env: | ||
| # Multiple API keys for fallback | ||
| GOOGLE_API_KEY: ${{ secrets.GOOGLE_API_KEY }} | ||
| GEMINI_API_KEY: ${{ secrets.GOOGLE_API_KEY }} | ||
| OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }} | ||
| ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }} | ||
| # Custom API URLs | ||
| GOOGLE_API_URL: ${{ secrets.GOOGLE_API_URL }} | ||
| ANTHROPIC_API_URL: ${{ secrets.ANTHROPIC_API_URL }} | ||
| ANTHROPIC_BASE_URL: ${{ secrets.ANTHROPIC_API_URL }} | ||
| OPENAI_API_URL: ${{ secrets.OPENAI_API_URL }} | ||
| LLM_BASE_URL: ${{ secrets.LLM_BASE_URL }} | ||
| # Model configuration | ||
| MODEL_NAME: ${{ matrix.ai_provider.model }} | ||
| FORCE_PROVIDER: ${{ secrets.FORCE_PROVIDER }} | ||
| GITHUB_CONTEXT: ${{ toJson(github) }} | ||
| continue-on-error: true | ||
| - name: 📊 Validate Incremental Test Results | ||
| if: steps.incremental_test.outcome == 'success' | ||
| run: | | ||
| echo "✅ Incremental Analysis completed successfully" | ||
| echo "Incremental Analysis Flag: ${{ steps.incremental_test.outputs.incremental-analysis }}" | ||
| # For synchronize events, incremental-analysis should be true | ||
| if [ "${{ github.event.action }}" == "synchronize" ] && [ "${{ steps.incremental_test.outputs.incremental-analysis }}" != "true" ]; then | ||
| echo "❌ Expected incremental analysis to be true for synchronize event" | ||
| exit 1 | ||
| fi | ||
| - name: 🧪 Test 3 - Comment Management (Check for Updates) | ||
| if: github.event_name == 'pull_request' && steps.xml_test.outcome == 'success' | ||
| run: | | ||
| echo "🔍 Testing comment management..." | ||
| # Use GitHub CLI to check if comments were updated rather than duplicated | ||
| COMMENT_COUNT=$(gh pr view ${{ github.event.number }} --json comments --jq '.comments | map(select(.body | contains("Visor Code Review"))) | length') | ||
| echo "Found $COMMENT_COUNT Visor review comments" | ||
| # Should have at most 1 review comment per PR (smart updating) | ||
| if [ "$COMMENT_COUNT" -gt 1 ]; then | ||
| echo "⚠️ Warning: Found $COMMENT_COUNT review comments. Smart updating may not be working properly." | ||
| else | ||
| echo "✅ Comment management working correctly" | ||
| fi | ||
| env: | ||
| GH_TOKEN: ${{ secrets.GITHUB_TOKEN }} | ||
| - name: 🧪 Test 4 - Different Check Types | ||
| if: matrix.ai_provider.api_key_secret != '' && secrets[matrix.ai_provider.api_key_secret] != '' | ||
| continue-on-error: true # Allow test to continue if quota issues occur | ||
| run: | | ||
| chmod +x ./dist/index.js | ||
| echo "🔍 Testing different check provider types..." | ||
| # Test all checks (which includes AI-powered review) | ||
| echo "Testing all checks with AI provider..." | ||
| ./dist/index.js --cli --check all --output json > ai_result.json | ||
| # Validate AI result | ||
| if ! jq -e '.summary.overallScore' ai_result.json > /dev/null; then | ||
| echo "❌ AI provider test failed" | ||
| exit 1 | ||
| fi | ||
| echo "✅ AI provider test passed" | ||
| echo "AI Result:" | ||
| jq '.summary' ai_result.json | ||
| env: | ||
| # Set the appropriate API key based on provider | ||
| GOOGLE_API_KEY: ${{ matrix.ai_provider.env_var == 'GOOGLE_API_KEY' && secrets.GOOGLE_API_KEY || '' }} | ||
| GEMINI_API_KEY: ${{ matrix.ai_provider.env_var == 'GOOGLE_API_KEY' && secrets.GOOGLE_API_KEY || '' }} | ||
| OPENAI_API_KEY: ${{ matrix.ai_provider.env_var == 'OPENAI_API_KEY' && secrets.OPENAI_API_KEY || '' }} | ||
| ANTHROPIC_API_KEY: ${{ matrix.ai_provider.env_var == 'ANTHROPIC_API_KEY' && secrets.ANTHROPIC_API_KEY || '' }} | ||
| # Custom API URLs | ||
| GOOGLE_API_URL: ${{ secrets.GOOGLE_API_URL }} | ||
| ANTHROPIC_API_URL: ${{ secrets.ANTHROPIC_API_URL }} | ||
| ANTHROPIC_BASE_URL: ${{ secrets.ANTHROPIC_API_URL }} | ||
| OPENAI_API_URL: ${{ secrets.OPENAI_API_URL }} | ||
| LLM_BASE_URL: ${{ secrets.LLM_BASE_URL }} | ||
| # Model configuration | ||
| MODEL_NAME: ${{ matrix.ai_provider.model }} | ||
| FORCE_PROVIDER: ${{ secrets.FORCE_PROVIDER }} | ||
| - name: 🧪 Test 5 - Output Format Validation | ||
| run: | | ||
| chmod +x ./dist/index.js | ||
| echo "🔍 Testing output formats..." | ||
| # Test table output (should not truncate messages) | ||
| echo "Testing table output formatting..." | ||
| ./dist/index.js --cli --check all --output table > table_output.txt | ||
| # Check that table contains expected sections | ||
| if grep -q "Analysis Summary" table_output.txt && grep -q "Issues" table_output.txt; then | ||
| echo "✅ Table output format test passed" | ||
| else | ||
| echo "❌ Table output format test failed" | ||
| cat table_output.txt | ||
| exit 1 | ||
| fi | ||
| # Test JSON output | ||
| echo "Testing JSON output..." | ||
| ./dist/index.js --cli --check all --output json > json_output.json | ||
| if jq -e '.summary' json_output.json > /dev/null; then | ||
| echo "✅ JSON output format test passed" | ||
| else | ||
| echo "❌ JSON output format test failed" | ||
| cat json_output.json | ||
| exit 1 | ||
| fi | ||
| - name: 🧪 Test 6 - Error Handling and Fallbacks | ||
| run: | | ||
| chmod +x ./dist/index.js | ||
| echo "🔍 Testing error handling..." | ||
| # Test with invalid API key (should gracefully handle) | ||
| echo "Testing invalid API key handling..." | ||
| # Clear all API keys and set an invalid one | ||
| unset GOOGLE_API_KEY OPENAI_API_KEY ANTHROPIC_API_KEY GEMINI_API_KEY | ||
| export GOOGLE_API_KEY="invalid-key-test" | ||
| ./dist/index.js --cli --check ai --output json > error_test.json || { | ||
| echo "✅ Error handling working - invalid API key properly rejected" | ||
| } | ||
| # Test with no API key (should use fallback if available) | ||
| echo "Testing fallback behavior..." | ||
| unset GOOGLE_API_KEY ANTHROPIC_API_KEY OPENAI_API_KEY GEMINI_API_KEY LLM_BASE_URL | ||
| ./dist/index.js --cli --check all --output json > fallback_test.json | ||
| if jq -e '.summary' fallback_test.json > /dev/null; then | ||
| echo "✅ Fallback behavior test passed" | ||
| else | ||
| echo "❌ Fallback behavior test failed" | ||
| exit 1 | ||
| fi | ||
| - name: 📈 Performance Benchmark | ||
| if: matrix.ai_provider.name == 'Google Gemini' && secrets.GOOGLE_API_KEY != '' | ||
| continue-on-error: true # Allow benchmark to continue if quota issues occur | ||
| run: | | ||
| chmod +x ./dist/index.js | ||
| echo "🚀 Running performance benchmark..." | ||
| start_time=$(date +%s%3N) | ||
| ./dist/index.js --cli --check all --output json > benchmark_result.json | ||
| end_time=$(date +%s%3N) | ||
| execution_time=$((end_time - start_time)) | ||
| echo "Execution time: ${execution_time}ms" | ||
| # Check if execution completed within reasonable time (60 seconds) | ||
| if [ $execution_time -lt 60000 ]; then | ||
| echo "✅ Performance benchmark passed (${execution_time}ms)" | ||
| else | ||
| echo "⚠️ Performance benchmark slow (${execution_time}ms)" | ||
| fi | ||
| env: | ||
| # Multiple API keys with fallback strategy | ||
| GOOGLE_API_KEY: ${{ secrets.GOOGLE_API_KEY }} | ||
| GEMINI_API_KEY: ${{ secrets.GOOGLE_API_KEY }} | ||
| OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }} | ||
| ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }} | ||
| # Custom API URLs for better quota management | ||
| GOOGLE_API_URL: ${{ secrets.GOOGLE_API_URL }} | ||
| ANTHROPIC_API_URL: ${{ secrets.ANTHROPIC_API_URL }} | ||
| ANTHROPIC_BASE_URL: ${{ secrets.ANTHROPIC_API_URL }} | ||
| OPENAI_API_URL: ${{ secrets.OPENAI_API_URL }} | ||
| LLM_BASE_URL: ${{ secrets.LLM_BASE_URL }} | ||
| # Model configuration | ||
| MODEL_NAME: ${{ matrix.ai_provider.model }} | ||
| FORCE_PROVIDER: ${{ secrets.FORCE_PROVIDER }} | ||
| - name: 📋 Test Results Summary | ||
| if: always() | ||
| run: | | ||
| echo "## 🧪 AI Battle Test Results Summary" | ||
| echo "**Provider:** ${{ matrix.ai_provider.name }}" | ||
| echo "**Event:** ${{ github.event_name }}${{ github.event.action && format(' ({0})', github.event.action) || '' }}" | ||
| echo "**Model:** ${{ matrix.ai_provider.model }}" | ||
| echo "" | ||
| echo "### Test Results:" | ||
| echo "- XML Formatted Analysis: ${{ steps.xml_test.outcome || 'skipped' }}" | ||
| echo "- Incremental Analysis: ${{ steps.incremental_test.outcome || 'skipped' }}" | ||
| echo "- Comment Management: ${{ github.event_name == 'pull_request' && 'tested' || 'skipped' }}" | ||
| echo "- Provider Types: ${{ matrix.ai_provider.api_key_secret != '' && 'tested' || 'skipped' }}" | ||
| echo "- Output Formats: tested" | ||
| echo "- Error Handling: tested" | ||
| echo "- Performance: ${{ matrix.ai_provider.name == 'Google Gemini' && 'tested' || 'skipped' }}" | ||
| echo "" | ||
| if [ "${{ steps.xml_test.outputs.review-score }}" ]; then | ||
| echo "### Metrics:" | ||
| echo "- Review Score: ${{ steps.xml_test.outputs.review-score }}/100" | ||
| echo "- Total Issues: ${{ steps.xml_test.outputs.total-issues }}" | ||
| echo "- Auto Review: ${{ steps.xml_test.outputs.auto-review-completed }}" | ||
| echo "- PR Action: ${{ steps.xml_test.outputs.pr-action }}" | ||
| echo "- Incremental: ${{ steps.xml_test.outputs.incremental-analysis }}" | ||
| fi | ||
| - name: 📤 Upload Test Artifacts | ||
| if: always() | ||
| uses: actions/upload-artifact@v4 | ||
| with: | ||
| name: ai-battle-test-results-${{ matrix.ai_provider.name }}-${{ github.event_name }} | ||
| path: | | ||
| *.json | ||
| *.txt | ||
| *.log | ||
| retention-days: 7 | ||
| battle-test-summary: | ||
| name: Battle Test Summary | ||
| needs: ai-battle-test | ||
| runs-on: ubuntu-latest | ||
| if: always() | ||
| steps: | ||
| - name: 📊 Generate Summary Report | ||
| run: | | ||
| echo "# 🤖 AI Battle Test Complete" | ||
| echo "" | ||
| echo "**Event:** ${{ github.event_name }}" | ||
| echo "**Trigger:** ${{ github.event.action || 'manual' }}" | ||
| echo "**Status:** ${{ needs.ai-battle-test.result }}" | ||
| echo "" | ||
| echo "## Test Matrix Results" | ||
| echo "All AI providers and scenarios have been tested." | ||
| echo "" | ||
| echo "Check individual job logs and artifacts for detailed results." | ||
