local-deep-research/.github/workflows/semgrep.yml

name: Semgrep Security Scan

on:
  workflow_call:  # Called by release-gate.yml
  workflow_dispatch:

permissions: {}  # Minimal top-level for OSSF Scorecard Token-Permissions

jobs:
  semgrep-scan:
    runs-on: ubuntu-latest
    timeout-minutes: 30
    permissions:
      contents: read
      security-events: write
      actions: read

    steps:
    - name: Harden the runner (Audit all outbound calls)
      uses: step-security/harden-runner@a5ad31d6a139d249332a2605b85202e8c0b78450 # v2.19.1
      with:
        egress-policy: audit

    - name: Checkout code
      uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2

      with:
        persist-credentials: false

    - name: Set up Python
      uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405 # v6.2.0
      with:
        python-version: '3.12'

    - name: Install Semgrep
      run: |
        # setuptools required: semgrep 1.87.0's opentelemetry dep imports pkg_resources,
        # which is not bundled with Python 3.12 by default on GitHub runners.
        # Pin setuptools<82 because 82.0 removed the pkg_resources module.
        pip install "setuptools<82" semgrep==1.87.0

    - name: Run Semgrep security scan
      run: |
        semgrep \
          --config=p/security-audit \
          --config=p/secrets \
          --severity=INFO \
          --json \
          --output=semgrep-results.json \
          src/ || true
        # Crash detection: if semgrep crashed, no output file was produced
        if [ ! -f semgrep-results.json ]; then
          echo "::error::Semgrep security scan crashed — no output produced"
          exit 1
        fi

    - name: Run Semgrep custom rules for LDR
      run: |
        semgrep \
          --config=.semgrep/rules/ \
          --severity=INFO \
          --json \
          --output=semgrep-custom-results.json \
          src/ || true
        # Crash detection: if semgrep crashed, no output file was produced
        if [ ! -f semgrep-custom-results.json ]; then
          echo "::error::Semgrep custom rules scan crashed — no output produced"
          exit 1
        fi

    - name: Merge Semgrep results
      run: |
        python3 -c "
        import json
        import glob

        results = []
        for file in glob.glob('semgrep-*.json'):
            try:
                with open(file) as f:
                    data = json.load(f)
                    if isinstance(data, dict) and 'results' in data:
                        results.extend(data['results'])
                    elif isinstance(data, list):
                        results.extend(data)
            except Exception as e:
                print(f'Error reading {file}: {e}')

        output = {'results': results, 'version': '1.0.0'}
        with open('semgrep-combined-results.json', 'w') as f:
            json.dump(output, f, indent=2)

        print(f'Combined {len(results)} findings from Semgrep scans')
        "

    - name: Convert to SARIF format
      run: |
        python3 -c "
        import json
        import uuid
        from datetime import datetime

        # Load combined results
        with open('semgrep-combined-results.json') as f:
            semgrep_data = json.load(f)

        # Convert to SARIF
        sarif = {
            '\$schema': 'https://json.schemastore.org/sarif-2.1.0',
            'version': '2.1.0',
            'runs': [{
                'tool': {
                    'driver': {
                        'name': 'Semgrep',
                        'version': '1.87.0',
                        'informationUri': 'https://semgrep.dev'
                    }
                },
                'results': []
            }]
        }

        for result in semgrep_data.get('results', []):
            sarif_result = {
                'ruleId': result.get('check_id', 'unknown'),
                'message': {
                    'text': result.get('message', 'Security issue detected')
                },
                'level': 'warning' if result.get('metadata', {}).get('severity', 'INFO') in ['ERROR', 'WARNING'] else 'note',
                'locations': [{
                    'physicalLocation': {
                        'artifactLocation': {
                            'uri': result.get('path', 'unknown')
                        },
                        'region': {
                            'startLine': result.get('start', {}).get('line', 1),
                            'startColumn': result.get('start', {}).get('col', 1),
                            'endLine': result.get('end', {}).get('line', result.get('start', {}).get('line', 1)),
                            'endColumn': result.get('end', {}).get('col', result.get('start', {}).get('col', 1) + 1)
                        }
                    }
                }]
            }

            # Add rule information
            metadata = result.get('metadata', {})
            sarif_result['rule'] = {
                'id': result.get('check_id', 'unknown'),
                'name': metadata.get('name', 'Security Issue'),
                'shortDescription': {
                    'text': metadata.get('name', 'Security Issue')
                },
                'fullDescription': {
                    'text': metadata.get('description', 'Security vulnerability detected')
                },
                'help': {
                    'text': metadata.get('remediation', 'Review and fix the security issue')
                },
                'properties': {
                    'precision': 'medium',
                    'tags': ['security', 'semgrep']
                }
            }

            if 'security-severity' in metadata:
                sarif_result['rule']['properties']['security-severity'] = metadata['security-severity']

            sarif['runs'][0]['results'].append(sarif_result)

        # Write SARIF file
        with open('semgrep-results.sarif', 'w') as f:
            json.dump(sarif, f, indent=2)

        print(f'Converted {len(sarif[\"runs\"][0][\"results\"])} findings to SARIF format')
        "

    - name: Ensure SARIF file exists
      if: always()
      run: |
        if [ ! -f semgrep-results.sarif ]; then
          echo "::warning::Semgrep SARIF conversion did not produce a file - creating empty placeholder"
          cat > semgrep-results.sarif << 'SARIF'
        {
          "version": "2.1.0",
          "$schema": "https://json.schemastore.org/sarif-2.1.0.json",
          "runs": [{
            "tool": {
              "driver": {
                "name": "Semgrep",
                "informationUri": "https://semgrep.dev",
                "rules": []
              }
            },
            "results": []
          }]
        }
        SARIF
        fi

    - name: Upload Semgrep results to GitHub Security tab
      uses: github/codeql-action/upload-sarif@68bde559dea0fdcac2102bfdf6230c5f70eb485e # v4.35.4
      if: always()
      with:
        sarif_file: 'semgrep-results.sarif'
        category: semgrep-security

    - name: Upload Semgrep results as artifact
      uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1
      if: always()
      with:
        name: semgrep-scan-results
        path: |
          semgrep-results.json
          semgrep-custom-results.json
          semgrep-combined-results.json
          semgrep-results.sarif
        retention-days: 7  # Reduced for security

    - name: Display Semgrep summary
      if: always()
      run: |
        {
          echo "## Semgrep Security Scan Summary"
          echo ""
        } >> "$GITHUB_STEP_SUMMARY"

        if [ -f semgrep-combined-results.json ]; then
          # Count results by severity
          CRITICAL=$(python3 -c "import json; data=json.load(open('semgrep-combined-results.json')); print(len([r for r in data['results'] if r.get('metadata', {}).get('severity') == 'ERROR']))" 2>/dev/null || echo "0")
          HIGH=$(python3 -c "import json; data=json.load(open('semgrep-combined-results.json')); print(len([r for r in data['results'] if r.get('metadata', {}).get('severity') == 'WARNING']))" 2>/dev/null || echo "0")
          MEDIUM=$(python3 -c "import json; data=json.load(open('semgrep-combined-results.json')); print(len([r for r in data['results'] if r.get('metadata', {}).get('severity') == 'INFO']))" 2>/dev/null || echo "0")
          TOTAL=$(python3 -c "import json; data=json.load(open('semgrep-combined-results.json')); print(len(data['results']))" 2>/dev/null || echo "0")

          {
            echo "📊 **Scan Results:**"
            echo "- **Critical:** $CRITICAL"
            echo "- **High:** $HIGH"
            echo "- **Medium:** $MEDIUM"
            echo "- **Total:** $TOTAL"
            echo ""
          } >> "$GITHUB_STEP_SUMMARY"

          if [ "$CRITICAL" -gt 0 ] || [ "$HIGH" -gt 0 ]; then
            echo "⚠️ **Action Required:** Critical or High severity issues found" >> "$GITHUB_STEP_SUMMARY"
          else
            echo "✅ **No Critical or High severity issues found**" >> "$GITHUB_STEP_SUMMARY"
          fi

          {
            echo ""
            echo "📋 **Detailed Results:**"
            echo "- Security tab: Results uploaded to GitHub Security tab"
            echo "- Artifacts: Full JSON and SARIF reports available"
          } >> "$GITHUB_STEP_SUMMARY"
        else
          echo "❌ Semgrep scan failed or no results generated" >> "$GITHUB_STEP_SUMMARY"
        fi