security: Fix CORS Configuration Vulnerabilities (CWE-942) - 3 instances of wildcard origins

[praisonai-triage-agent]

Security Vulnerability Report: CORS Misconfiguration (CWE-942)

Background Context

Following the security audit conducted for PR #1319, 3 instances of CORS misconfiguration vulnerabilities remain unaddressed across the PraisonAI codebase. These were identified by semgrep security scanning using the OWASP Top 10 and Python security rulesets.

Related Work: PR #1319 successfully fixed 29 MD5/debug vulnerabilities, but deliberately excluded these CORS issues as they require architectural decisions from maintainers.

Vulnerability Summary

Category	Count	CWE	Risk Level
Wildcard CORS allow_origins=["*"]	3	CWE-942	MEDIUM-HIGH

Technical Details

Wildcard CORS Origins (3 instances)

Configuration: allow_origins=["*"] in CORS middleware setup
Risk: Any origin can make cross-origin requests to the API
Impact:

• Credential theft via malicious websites
• CSRF attacks bypassing same-origin policy
• Data exfiltration from authenticated sessions
• Cross-domain request forgery

Example Vulnerable Pattern:

# Vulnerable - allows any origin
app.add_middleware(
    CORSMiddleware,
    allow_origins=["*"],  # SECURITY RISK
    allow_credentials=True,
    allow_methods=["*"],
    allow_headers=["*"]
)

Security Implications

1. Cross-Origin Resource Sharing Bypass

• Malicious websites can make authenticated requests to PraisonAI APIs
• User credentials and session tokens accessible from any domain
• Violates same-origin policy protection

2. Data Exfiltration Risk

• Sensitive data (agent configurations, chat history, API keys) exposed
• Cross-domain data theft through authenticated requests
• Privacy violations for multi-tenant deployments

3. CSRF Attack Vector

• Cross-site request forgery becomes trivial with wildcard origins
• Malicious sites can perform actions on behalf of authenticated users
• Bypasses CSRF protections that rely on origin validation

Recommended Remediation Strategy

Phase 1: Immediate Security Hardening

1. Replace wildcard origins with explicit allow-lists:

# Secure configuration
ALLOWED_ORIGINS = [
    "http://localhost:3000",     # Development frontend
    "http://localhost:8000",     # Local development
    "https://app.praisonai.com",  # Production frontend
    # Add other trusted domains as needed
]

app.add_middleware(
    CORSMiddleware,
    allow_origins=ALLOWED_ORIGINS,  # Explicit origins only
    allow_credentials=True,
    allow_methods=["GET", "POST", "PUT", "DELETE"],  # Restrict methods
    allow_headers=["Authorization", "Content-Type"],     # Restrict headers
)

2. Environment-based configuration:

import os

# Production vs development origins
if os.getenv("ENVIRONMENT") == "production":
    ALLOWED_ORIGINS = ["https://app.praisonai.com"]
else:
    ALLOWED_ORIGINS = [
        "http://localhost:3000",
        "http://localhost:8000",
        "http://127.0.0.1:3000",
    ]

Phase 2: Enhanced Security Controls

3. Origin validation middleware:

def validate_origin(request):
    origin = request.headers.get("Origin")
    if origin and origin not in ALLOWED_ORIGINS:
        raise HTTPException(403, "Origin not allowed")
    return origin

4. Dynamic origin configuration:

# Allow runtime configuration via environment variables
CORS_ORIGINS = os.getenv("CORS_ALLOWED_ORIGINS", "").split(",")
ALLOWED_ORIGINS = [origin.strip() for origin in CORS_ORIGINS if origin.strip()]

Phase 3: Security Monitoring

5. CORS request logging:

# Log cross-origin requests for monitoring
@app.middleware("http")
async def log_cors_requests(request, call_next):
    origin = request.headers.get("Origin")
    if origin:
        logger.info(f"CORS request from origin: {origin}")
    return await call_next(request)

6. Origin validation alerts:

# Alert on suspicious origin requests
def check_suspicious_origins(origin):
    suspicious_patterns = ["localhost", "127.0.0.1", "file://"]
    if any(pattern in origin for pattern in suspicious_patterns):
        logger.warning(f"Suspicious CORS origin: {origin}")

Deployment Considerations

Development vs Production

• Development: Allow localhost origins for development convenience
• Staging: Restrict to staging domain only
• Production: Strict allow-list of production domains only

Multi-tenant Deployments

• Dynamic origin configuration per tenant
• Tenant-specific CORS policies
• Origin validation against tenant domain registry

CDN and Proxy Configurations

• Account for CDN domains in origin allow-lists
• Handle forwarded origin headers correctly
• Validate against actual client origins, not proxy origins

Affected Components

Likely locations of CORS configurations:

• Main API Server: FastAPI/Flask application setup
• Agent API Endpoints: WebSocket and HTTP APIs
• Development Servers: Local development configurations
• CLI Web Interface: Browser-based tools and dashboards

Implementation Steps

1. Audit CORS configurations:

# Find all CORS middleware configurations
grep -r "allow_origins" . --include="*.py"
grep -r "CORSMiddleware" . --include="*.py" 
grep -r "CORS" . --include="*.py"

2. Environment-specific configuration:

# Create environment-specific CORS settings
echo "CORS_ALLOWED_ORIGINS=https://app.praisonai.com" >> .env.production
echo "CORS_ALLOWED_ORIGINS=http://localhost:3000,http://localhost:8000" >> .env.development

3. Validate configuration:

# Test CORS policy enforcement
curl -H "Origin: https://evil.com" -I http://localhost:8000/api/health
# Should return 403 or omit CORS headers

Security Testing

CORS Policy Validation:

# Test legitimate origins
curl -H "Origin: https://app.praisonai.com" -I http://api.praisonai.com/health

# Test malicious origins  
curl -H "Origin: https://malicious.com" -I http://api.praisonai.com/health

# Test credentials with cross-origin request
curl -H "Origin: https://evil.com" -H "Cookie: session=abc123" -I http://api.praisonai.com/api/user

Automated Security Testing:

# Test CORS policy enforcement
def test_cors_policy():
    # Should allow configured origins
    response = requests.get(url, headers={"Origin": "https://app.praisonai.com"})
    assert "Access-Control-Allow-Origin" in response.headers
    
    # Should reject unconfigured origins  
    response = requests.get(url, headers={"Origin": "https://evil.com"})
    assert "Access-Control-Allow-Origin" not in response.headers

References

• CWE-942: Permissive Cross-domain Policy with Untrusted Domains - https://cwe.mitre.org/data/definitions/942.html
• OWASP: Cross-Origin Resource Sharing (CORS) - https://owasp.org/www-community/attacks/CORS_OriginHeaderScrutiny
• MDN: Cross-Origin Resource Sharing - https://developer.mozilla.org/en-US/docs/Web/HTTP/CORS

Success Criteria

• All 3 wildcard CORS configurations replaced with explicit allow-lists
• Environment-specific CORS policies implemented
• CORS policy validation added to test suite
• Security scan shows 0 CWE-942 findings
• Documentation updated with secure CORS configuration guidelines

---

Priority: MEDIUM-HIGH | Effort: Small to Medium
Dependencies: Frontend domain configuration, deployment environment setup

This issue tracks the remediation of CORS misconfiguration vulnerabilities identified during the security audit conducted for PR #1319.

[praisonai-triage-agent]

👋 Thanks for opening this issue, @praisonai-triage-agent[bot]!

A maintainer will review this shortly. In the meantime:

• Make sure you've included steps to reproduce (for bugs)
• Check existing issues for duplicates
• Review the documentation for related guides

A maintainer can trigger deeper analysis by commenting @claude on this issue.

[praisonai-triage-agent]

Claude finished @praisonai-triage-agent[bot]'s task —— View job • claude/issue-1321-20260408-2100 • [Create PR ➔](https://github.com/MervinPraison/PraisonAI/compare/main...claude/issue-1321-20260408-2100?quick_pull=1&title=Issue%20%231321%3A%20Changes%20from%20Claude&body=This%20PR%20addresses%20issue%20%231321%0A%0AGenerated%20with%20%5BClaude%20Code%5D(https%3A%2F%2Fclaude.ai%2Fcode)

---

✅ CORS Configuration Vulnerabilities Fixed (CWE-942)

I have successfully addressed all 3 instances of CORS misconfiguration vulnerabilities identified in the security audit. All wildcard CORS origins (allow_origins=["*"]) have been replaced with secure, environment-based configurations.

🔍 Vulnerabilities Found & Fixed:

1. src/praisonai/praisonai/browser/server.py:88 ✅ FIXED

   • Before: allow_origins=["*"] (hardcoded wildcard)
   • After: Environment-based BROWSER_CORS_ORIGINS with secure localhost + Chrome extension defaults

2. src/praisonai/praisonai/jobs/server.py:90 ✅ FIXED

   • Before: cors_origins or ["*"] (wildcard fallback)
   • After: Environment-based JOBS_CORS_ORIGINS with production-safe defaults (empty in production)

3. examples/python/api/secondary-market-research-api.py:49 ✅ FIXED

   • Before: allow_origins=["*"] (hardcoded wildcard)
   • After: Environment-based API_CORS_ORIGINS with localhost-only defaults

🛡️ Additional Security Improvements:

• Config Class Defaults: Removed wildcard defaults from ServerConfig, GatewayConfig, and AppConfig
• MCP Server Transport: Replaced wildcard with secure localhost origins
• CORS Headers & Methods: Restricted from ["*"] to essential headers and methods only
• Tests & Documentation: Updated to reflect secure defaults

✅ Todo List:

• Set up git identity and authentication
• Read AGENTS.md for architecture guidelines
• Search for CORS configurations in the codebase
• Fix browser/server.py - replace wildcard CORS with secure config
• Fix jobs/server.py - replace default wildcard with environment-based config
• Fix example API - replace wildcard with proper config
• Fix additional wildcard defaults in config classes
• Update tests and documentation
• Test changes
• Commit and push changes
• Create pull request

🚀 Pull Request Created:

PR #1323: security: Fix CORS Configuration Vulnerabilities (CWE-942)

🔐 Security Configuration:

Environment Variables (for production):

export BROWSER_CORS_ORIGINS="https://app.praisonai.com"
export JOBS_CORS_ORIGINS="https://app.praisonai.com,https://admin.praisonai.com" 
export API_CORS_ORIGINS="https://app.praisonai.com"
export ENVIRONMENT="production"

Development (automatic localhost defaults):

• http://localhost:3000, http://localhost:8000, http://127.0.0.1:3000, http://127.0.0.1:8000
• Chrome extension origins: chrome-extension://*

✅ Verification:

• No remaining allow_origins=["*"] found in codebase
• All configuration classes default to empty CORS origins
• Unit tests updated and passing
• Environment-based configuration working correctly

Security Impact: This fix eliminates all CWE-942 vulnerabilities by preventing arbitrary cross-origin requests, blocking CSRF attacks, and enforcing proper same-origin policy protection.

--- | Branch: claude/issue-1321-20260408-2100 | Pull Request #1323