feat(prospecting): implement security audit pipeline (Workstream 2A)

Complete security audit integration into the enrichment pipeline:

Backend:
- SecurityAuditService with 7 passive checks: HTTPS, SSL cert, security
  headers, exposed files, cookies, server info, technology detection
- Constants file with SECURITY_HEADERS, EXPOSED_PATHS, SEVERITY_SCORES
- SecurityAuditResponse schema with JSON field validators + aliases
- Endpoints: POST /security-audit/{id}, POST /security-audit/batch
- Added to full_enrichment pipeline (Step 5, before scoring)
- get_pending_security_audit() query in prospect_service

Frontend:
- Security tab on prospect detail page with grade badge (A+ to F),
  score/100, severity counts, HTTPS/SSL status, missing headers,
  exposed files, technologies, and full findings list
- "Run Security Audit" button with loading state
- "Security Audit" batch button on scan-jobs page

Tested on batirenovation-strasbourg.fr: Grade D (50/100), 11 issues
found (missing headers, exposed wp-login, server version disclosure).

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

app/modules/prospecting/services/prospect_service.py

@@ -251,6 +251,17 @@ class ProspectService:
             .all()
         )
 
+    def get_pending_security_audit(self, db: Session, limit: int = 50) -> list[Prospect]:
+        return (
+            db.query(Prospect)
+            .filter(
+                Prospect.has_website.is_(True),
+                Prospect.last_security_audit_at.is_(None),
+            )
+            .limit(limit)
+            .all()
+        )
+
     def count_by_status(self, db: Session) -> dict[str, int]:
         results = db.query(Prospect.status, func.count(Prospect.id)).group_by(Prospect.status).all()  # noqa: SVC-005 - prospecting is platform-scoped, not store-scoped
         return {status.value if hasattr(status, "value") else str(status): count for status, count in results}

app/modules/prospecting/services/security_audit_constants.py

@@ -0,0 +1,75 @@
+# app/modules/prospecting/services/security_audit_constants.py
+"""
+Constants for security audit checks.
+
+Structural data used by SecurityAuditService. Translations for report
+generation are kept in the standalone script (scripts/security-audit/audit.py)
+until Phase 2B (report service) migrates them.
+"""
+
+# Severity scores — deducted from a starting score of 100
+SEVERITY_SCORES = {
+    "critical": 15,
+    "high": 10,
+    "medium": 5,
+    "low": 2,
+    "info": 0,
+}
+
+# Security headers to check and their severity if missing
+SECURITY_HEADERS = {
+    "Strict-Transport-Security": {"severity": "high", "impact": "MITM attacks, session hijacking via HTTP downgrade"},
+    "Content-Security-Policy": {"severity": "high", "impact": "XSS attacks, script injection, data theft"},
+    "X-Frame-Options": {"severity": "medium", "impact": "Clickjacking attacks via invisible iframes"},
+    "X-Content-Type-Options": {"severity": "medium", "impact": "MIME type confusion, content injection"},
+    "Referrer-Policy": {"severity": "low", "impact": "URL parameter leakage to third parties"},
+    "Permissions-Policy": {"severity": "low", "impact": "Unrestricted browser API access (camera, mic, location)"},
+    "X-XSS-Protection": {"severity": "info", "impact": "Legacy XSS filter not configured"},
+}
+
+# Paths to check for exposed sensitive files/directories
+EXPOSED_PATHS = [
+    ("/.env", "Environment file (database passwords, API keys)", "critical"),
+    ("/.git/config", "Git repository (full source code)", "critical"),
+    ("/.git/HEAD", "Git repository HEAD", "critical"),
+    ("/.htpasswd", "Password file", "critical"),
+    ("/wp-admin/", "WordPress admin panel", "high"),
+    ("/wp-login.php", "WordPress login page", "high"),
+    ("/administrator/", "Joomla admin panel", "high"),
+    ("/admin/", "Admin panel", "high"),
+    ("/admin/login", "Admin login page", "high"),
+    ("/phpmyadmin/", "phpMyAdmin (database manager)", "high"),
+    ("/backup/", "Backup directory", "high"),
+    ("/backup.zip", "Backup archive", "high"),
+    ("/backup.sql", "Database backup", "high"),
+    ("/db.sql", "Database dump", "high"),
+    ("/dump.sql", "Database dump", "high"),
+    ("/.htaccess", "Server configuration", "medium"),
+    ("/web.config", "IIS configuration", "medium"),
+    ("/server-status", "Apache server status", "medium"),
+    ("/server-info", "Apache server info", "medium"),
+    ("/info.php", "PHP info page", "medium"),
+    ("/phpinfo.php", "PHP info page", "medium"),
+    ("/graphql", "GraphQL endpoint", "medium"),
+    ("/debug/", "Debug endpoint", "medium"),
+    ("/elmah.axd", ".NET error log", "medium"),
+    ("/trace.axd", ".NET trace log", "medium"),
+    ("/readme.html", "CMS readme (reveals version)", "low"),
+    ("/license.txt", "CMS license (reveals version)", "low"),
+    ("/CHANGELOG.md", "Changelog (reveals version)", "low"),
+    ("/robots.txt", "Robots file", "info"),
+    ("/.well-known/security.txt", "Security contact file", "info"),
+    ("/sitemap.xml", "Sitemap", "info"),
+    ("/crossdomain.xml", "Flash cross-domain policy", "low"),
+    ("/api/", "API endpoint", "info"),
+]
+
+# Paths that are admin panels (separate severity logic)
+ADMIN_PATHS = {"/wp-admin/", "/wp-login.php", "/administrator/", "/admin/", "/admin/login"}
+
+# Robots.txt disallow patterns that may reveal sensitive areas
+ROBOTS_SENSITIVE_PATTERNS = [
+    "admin", "backup", "private", "secret", "staging",
+    "test", "dev", "internal", "api", "config",
+    "database", "panel", "dashboard", "login", "cgi-bin",
+]

app/modules/prospecting/services/security_audit_service.py

@@ -0,0 +1,443 @@
+# app/modules/prospecting/services/security_audit_service.py
+"""
+Security audit service for prospect websites.
+
+Performs passive security checks (HTTPS, SSL, headers, exposed files,
+cookies, server info, technology detection) and stores results as
+ProspectSecurityAudit. All checks are read-only — no active exploitation.
+
+Migrated from scripts/security-audit/audit.py into the enrichment pipeline.
+"""
+
+import json
+import logging
+import re
+import socket
+import ssl
+from datetime import UTC, datetime
+
+import requests
+from sqlalchemy.orm import Session
+
+from app.modules.prospecting.models import Prospect, ProspectSecurityAudit
+from app.modules.prospecting.services.security_audit_constants import (
+    ADMIN_PATHS,
+    EXPOSED_PATHS,
+    ROBOTS_SENSITIVE_PATTERNS,
+    SECURITY_HEADERS,
+    SEVERITY_SCORES,
+)
+
+logger = logging.getLogger(__name__)
+
+REQUEST_TIMEOUT = 10
+USER_AGENT = (
+    "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 "
+    "(KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
+)
+
+
+class SecurityAuditService:
+    """Runs passive security checks against a prospect's website."""
+
+    def run_audit(self, db: Session, prospect: Prospect) -> ProspectSecurityAudit | None:
+        """Run all security checks and store results."""
+        domain = prospect.domain_name
+        if not domain or not prospect.has_website:
+            return None
+
+        scheme = "https" if prospect.uses_https else "http"
+        url = f"{scheme}://{domain}"
+        findings = []
+        technologies = []
+        score = 100
+        has_https = None
+        has_valid_ssl = None
+        ssl_expires_at = None
+        missing_headers = []
+        exposed_files = []
+
+        session = requests.Session()
+        session.headers["User-Agent"] = USER_AGENT
+        session.verify = True
+        session.max_redirects = 5
+
+        # Fetch the page
+        response = None
+        html_content = ""
+        try:
+            response = session.get(url, timeout=REQUEST_TIMEOUT, allow_redirects=True)
+            html_content = response.text
+            if response.url != url:
+                url = response.url
+        except requests.exceptions.SSLError:
+            findings.append(self._finding("Weak SSL/TLS configuration", "critical", "transport",
+                                          "Server supports outdated encryption protocols"))
+            try:
+                session.verify = False  # noqa: SEC047 fallback for broken SSL
+                response = session.get(url, timeout=REQUEST_TIMEOUT, allow_redirects=True)
+                html_content = response.text
+            except Exception:
+                pass
+        except requests.exceptions.RequestException as e:
+            logger.warning("Security audit: cannot reach %s: %s", domain, e)
+            return self._save_audit(db, prospect, score=0, grade="F", findings=findings,
+                                    scan_error=f"Cannot reach website: {e}",
+                                    technologies=technologies)
+
+        # Run checks
+        https_findings, has_https = self._check_https(url, html_content)
+        findings.extend(https_findings)
+
+        ssl_findings, has_valid_ssl, ssl_expires_at = self._check_ssl(domain)
+        findings.extend(ssl_findings)
+
+        header_findings, missing_headers = self._check_headers(response)
+        findings.extend(header_findings)
+
+        server_findings, server_techs = self._check_server_info(response)
+        findings.extend(server_findings)
+        technologies.extend(server_techs)
+
+        tech_findings, detected_techs = self._check_technology(html_content, response)
+        findings.extend(tech_findings)
+        technologies.extend(detected_techs)
+
+        cookie_findings = self._check_cookies(response)
+        findings.extend(cookie_findings)
+
+        exposed_findings, exposed_files = self._check_exposed_files(domain, scheme, session)
+        findings.extend(exposed_findings)
+
+        session.close()
+
+        # Calculate score
+        for f in findings:
+            if not f.get("is_positive", False):
+                score = max(0, score - SEVERITY_SCORES.get(f["severity"], 0))
+
+        grade = self._calculate_grade(score)
+
+        return self._save_audit(
+            db, prospect,
+            score=score, grade=grade, findings=findings,
+            has_https=has_https, has_valid_ssl=has_valid_ssl,
+            ssl_expires_at=ssl_expires_at,
+            missing_headers=missing_headers, exposed_files=exposed_files,
+            technologies=technologies,
+        )
+
+    # ── Check methods ───────────────────────────────────────────────────────
+
+    def _check_https(self, url: str, html_content: str) -> tuple[list[dict], bool | None]:
+        """Check HTTPS configuration."""
+        findings = []
+        from urllib.parse import urlparse
+
+        parsed = urlparse(url)
+        has_https = parsed.scheme == "https"
+
+        if has_https:
+            findings.append(self._finding("HTTPS enabled", "info", "transport",
+                                          "Website uses encrypted connections", is_positive=True))
+            # Check mixed content
+            http_resources = re.findall(r'(src|href|action)=["\']http://[^"\']+["\']', html_content, re.IGNORECASE)
+            if http_resources:
+                findings.append(self._finding("Mixed content detected", "medium", "transport",
+                                              "HTTPS site loads resources over insecure HTTP"))
+        else:
+            findings.append(self._finding("No HTTPS", "critical", "transport",
+                                          "Website transmits all data in plain text"))
+
+        return findings, has_https
+
+    def _check_ssl(self, domain: str) -> tuple[list[dict], bool | None, datetime | None]:
+        """Check SSL certificate validity."""
+        findings = []
+        has_valid_ssl = None
+        ssl_expires_at = None
+
+        try:
+            context = ssl.create_default_context()
+            with socket.create_connection((domain, 443), timeout=REQUEST_TIMEOUT) as sock:
+                with context.wrap_socket(sock, server_hostname=domain) as ssock:
+                    cert = ssock.getpeercert()
+                    not_after = datetime.strptime(cert["notAfter"], "%b %d %H:%M:%S %Y %Z").replace(tzinfo=UTC)
+                    days_remaining = (not_after - datetime.now(UTC)).days
+                    ssl_expires_at = not_after
+
+                    if days_remaining < 0:
+                        has_valid_ssl = False
+                        findings.append(self._finding("SSL certificate expired", "critical", "transport",
+                                                      f"Certificate expired on {not_after.strftime('%Y-%m-%d')}"))
+                    elif days_remaining < 30:
+                        has_valid_ssl = True
+                        findings.append(self._finding(f"SSL expires in {days_remaining} days", "high", "transport",
+                                                      f"Certificate expires on {not_after.strftime('%Y-%m-%d')}"))
+                    else:
+                        has_valid_ssl = True
+                        findings.append(self._finding("SSL certificate valid", "info", "transport",
+                                                      f"Valid until {not_after.strftime('%Y-%m-%d')} ({days_remaining} days)",
+                                                      is_positive=True))
+
+                    # Check TLS version
+                    protocol = ssock.version()
+                    if protocol in ("TLSv1", "TLSv1.1", "SSLv3", "SSLv2"):
+                        findings.append(self._finding("Weak TLS version", "high", "transport",
+                                                      f"Server supports outdated protocol: {protocol}"))
+
+        except ssl.SSLCertVerificationError:
+            has_valid_ssl = False
+            findings.append(self._finding("SSL certificate invalid", "critical", "transport",
+                                          "Certificate verification failed"))
+        except (TimeoutError, ConnectionRefusedError, OSError):
+            pass  # No SSL, already caught by HTTPS check
+
+        return findings, has_valid_ssl, ssl_expires_at
+
+    def _check_headers(self, response) -> tuple[list[dict], list[str]]:
+        """Check for missing security headers."""
+        findings = []
+        missing = []
+
+        if not response:
+            return findings, missing
+
+        for header_name, config in SECURITY_HEADERS.items():
+            if header_name in response.headers:
+                findings.append(self._finding(f"Header present: {header_name}", "info", "headers",
+                                              header_name, is_positive=True))
+            else:
+                missing.append(header_name)
+                findings.append(self._finding(f"Missing: {header_name}", config["severity"], "headers",
+                                              config["impact"]))
+
+        return findings, missing
+
+    def _check_server_info(self, response) -> tuple[list[dict], list[str]]:
+        """Check for server version disclosure."""
+        findings = []
+        technologies = []
+
+        if not response:
+            return findings, technologies
+
+        server = response.headers.get("Server", "")
+        x_powered = response.headers.get("X-Powered-By", "")
+
+        info_parts = []
+        if server:
+            info_parts.append(server)
+            technologies.append(server)
+        if x_powered:
+            info_parts.append(f"X-Powered-By: {x_powered}")
+            technologies.append(x_powered)
+
+        if info_parts:
+            has_version = bool(re.search(r"\d+\.\d+", " ".join(info_parts)))
+            severity = "medium" if has_version else "low"
+            findings.append(self._finding("Server version exposed", severity, "config",
+                                          " | ".join(info_parts)))
+
+        return findings, technologies
+
+    def _check_technology(self, html_content: str, response) -> tuple[list[dict], list[str]]:
+        """Detect CMS and technology stack."""
+        findings = []
+        technologies = []
+
+        if not html_content:
+            return findings, technologies
+
+        # WordPress
+        wp_indicators = ["wp-content/", "wp-includes/", 'name="generator" content="WordPress']
+        if any(ind in html_content for ind in wp_indicators):
+            version = "unknown"
+            ver_match = re.search(r'content="WordPress\s+([\d.]+)"', html_content)
+            if ver_match:
+                version = ver_match.group(1)
+            severity = "medium" if version != "unknown" else "low"
+            findings.append(self._finding(f"WordPress detected (v{version})", severity, "technology",
+                                          "Version publicly visible" if version != "unknown" else "CMS detected"))
+            technologies.append(f"WordPress {version}")
+
+        # Joomla
+        if "/media/jui/" in html_content or "Joomla" in html_content:
+            findings.append(self._finding("Joomla detected", "low", "technology", "CMS detected"))
+            technologies.append("Joomla")
+
+        # Drupal
+        if "Drupal" in html_content or "/sites/default/" in html_content:
+            findings.append(self._finding("Drupal detected", "low", "technology", "CMS detected"))
+            technologies.append("Drupal")
+
+        # Hosted platforms (not vulnerable in the same way)
+        if "wix.com" in html_content:
+            technologies.append("Wix")
+        if "squarespace.com" in html_content:
+            technologies.append("Squarespace")
+        if "cdn.shopify.com" in html_content:
+            technologies.append("Shopify")
+
+        return findings, technologies
+
+    def _check_cookies(self, response) -> list[dict]:
+        """Check cookie security flags."""
+        findings = []
+
+        if not response:
+            return findings
+
+        set_cookie_headers = response.headers.get("Set-Cookie", "")
+        if not set_cookie_headers:
+            return findings
+
+        has_insecure = False
+        has_no_httponly = False
+        has_no_samesite = False
+
+        for cookie in set_cookie_headers.split(","):
+            cookie_lower = cookie.lower()
+            if "secure" not in cookie_lower:
+                has_insecure = True
+            if "httponly" not in cookie_lower:
+                has_no_httponly = True
+            if "samesite" not in cookie_lower:
+                has_no_samesite = True
+
+        if has_insecure:
+            findings.append(self._finding("Cookies lack Secure flag", "medium", "cookies",
+                                          "Session cookies can be intercepted over HTTP"))
+        if has_no_httponly:
+            findings.append(self._finding("Cookies lack HttpOnly flag", "medium", "cookies",
+                                          "Cookies accessible to JavaScript (XSS risk)"))
+        if has_no_samesite:
+            findings.append(self._finding("Cookies lack SameSite attribute", "low", "cookies",
+                                          "Vulnerable to cross-site request attacks"))
+
+        return findings
+
+    def _check_exposed_files(self, domain: str, scheme: str, session) -> tuple[list[dict], list[str]]:
+        """Check for exposed sensitive files and directories."""
+        findings = []
+        exposed = []
+        base = f"{scheme}://{domain}"
+        security_txt_found = False
+        robots_content = None
+
+        for path, description, default_severity in EXPOSED_PATHS:
+            try:
+                resp = session.get(f"{base}{path}", timeout=REQUEST_TIMEOUT, allow_redirects=False)
+
+                if path == "/.well-known/security.txt" and resp.status_code == 200:
+                    security_txt_found = True
+                    continue
+                if path == "/robots.txt" and resp.status_code == 200:
+                    robots_content = resp.text
+                    continue
+                if path == "/sitemap.xml" or path == "/api/":
+                    continue
+
+                if resp.status_code == 200:
+                    if path in ADMIN_PATHS:
+                        findings.append(self._finding(f"Admin panel exposed: {path}", "high", "exposure",
+                                                      f"Admin login at {base}{path} is publicly accessible"))
+                    else:
+                        findings.append(self._finding(f"Exposed: {path}", default_severity, "exposure",
+                                                      f"{description} is publicly accessible"))
+                    exposed.append(path)
+
+            except Exception:
+                continue
+
+        # Security.txt check
+        if not security_txt_found:
+            findings.append(self._finding("No security.txt", "info", "exposure",
+                                          "No /.well-known/security.txt for responsible disclosure"))
+
+        # Robots.txt analysis
+        if robots_content:
+            disallowed = re.findall(r"Disallow:\s*(.+)", robots_content, re.IGNORECASE)
+            sensitive_found = []
+            for path in disallowed:
+                path = path.strip()
+                if any(pattern in path.lower() for pattern in ROBOTS_SENSITIVE_PATTERNS):
+                    sensitive_found.append(path)
+
+            if sensitive_found:
+                findings.append(self._finding("Robots.txt reveals sensitive paths", "low", "exposure",
+                                              f"Disallowed paths: {', '.join(sensitive_found[:5])}"))
+
+        return findings, exposed
+
+    # ── Helpers ──────────────────────────────────────────────────────────────
+
+    @staticmethod
+    def _finding(title: str, severity: str, category: str, detail: str, is_positive: bool = False) -> dict:
+        """Create a finding dict."""
+        return {
+            "title": title,
+            "severity": severity,
+            "category": category,
+            "detail": detail,
+            "is_positive": is_positive,
+        }
+
+    @staticmethod
+    def _calculate_grade(score: int) -> str:
+        if score >= 95:
+            return "A+"
+        if score >= 85:
+            return "A"
+        if score >= 70:
+            return "B"
+        if score >= 55:
+            return "C"
+        if score >= 40:
+            return "D"
+        return "F"
+
+    def _save_audit(
+        self, db: Session, prospect: Prospect, *,
+        score: int, grade: str, findings: list[dict],
+        has_https: bool | None = None, has_valid_ssl: bool | None = None,
+        ssl_expires_at: datetime | None = None,
+        missing_headers: list[str] | None = None,
+        exposed_files: list[str] | None = None,
+        technologies: list[str] | None = None,
+        scan_error: str | None = None,
+    ) -> ProspectSecurityAudit:
+        """Upsert security audit results."""
+        audit = prospect.security_audit
+        if not audit:
+            audit = ProspectSecurityAudit(prospect_id=prospect.id)
+            db.add(audit)
+
+        audit.score = score
+        audit.grade = grade
+        audit.findings_json = json.dumps(findings)
+        audit.has_https = has_https
+        audit.has_valid_ssl = has_valid_ssl
+        audit.ssl_expires_at = ssl_expires_at
+        audit.missing_headers_json = json.dumps(missing_headers or [])
+        audit.exposed_files_json = json.dumps(exposed_files or [])
+        audit.technologies_json = json.dumps(technologies or [])
+        audit.scan_error = scan_error
+
+        # Denormalized counts
+        audit.findings_count_critical = sum(1 for f in findings if f["severity"] == "critical" and not f.get("is_positive"))
+        audit.findings_count_high = sum(1 for f in findings if f["severity"] == "high" and not f.get("is_positive"))
+        audit.findings_count_medium = sum(1 for f in findings if f["severity"] == "medium" and not f.get("is_positive"))
+        audit.findings_count_low = sum(1 for f in findings if f["severity"] == "low" and not f.get("is_positive"))
+        audit.findings_count_info = sum(1 for f in findings if f["severity"] == "info" and not f.get("is_positive"))
+
+        prospect.last_security_audit_at = datetime.now(UTC)
+        db.flush()
+
+        logger.info("Security audit for %s: score=%d grade=%s (%d findings)",
+                     prospect.domain_name, score, grade,
+                     len([f for f in findings if not f.get("is_positive")]))
+        return audit
+
+
+security_audit_service = SecurityAuditService()