We release patches for security vulnerabilities in the following versions:

We take the security of LionAGI QE Fleet seriously. If you discover a security vulnerability, please follow these steps:

Do not open a public GitHub issue for security vulnerabilities. This helps protect users who haven't yet updated.

Email security concerns to: security@agentic-qe.dev

Please include:

• Description of the vulnerability
• Steps to reproduce the issue
• Potential impact assessment
• Affected versions
• Suggested fix (if any)
• Your contact information for follow-up

• Initial Response: Within 48 hours
• Status Updates: Every 5 business days
• Fix Timeline: Critical issues within 7 days, others within 30 days
• Public Disclosure: After patch release and coordination with reporter

Security researchers who responsibly disclose vulnerabilities will be:

• Credited in the CHANGELOG.md (unless anonymity is requested)
• Listed in our security acknowledgments
• Eligible for our bug bounty program (if applicable)

When using LionAGI QE Fleet, follow these security guidelines:

# ✅ GOOD: Validate all external inputs
def execute_test(file_path: str):
    if not os.path.abspath(file_path).startswith(PROJECT_ROOT):
        raise ValueError("Invalid file path")
    # Safe to proceed

# ❌ BAD: No validation
def execute_test(file_path: str):
    subprocess.run(f"pytest {file_path}", shell=True)  # VULNERABLE!

# ✅ GOOD: Use list arguments, disable shell
subprocess.run(
    ["pytest", file_path, "-v"],
    shell=False,
    timeout=60
)

# ❌ BAD: String concatenation with shell=True
subprocess.run(
    f"pytest {file_path} -v",
    shell=True  # Command injection risk!
)

# ✅ GOOD: Use JSON for data serialization
import json
data = json.loads(input_data)

# ❌ BAD: Never use pickle for untrusted data
import pickle
data = pickle.loads(input_data)  # Arbitrary code execution!

# ✅ GOOD: Whitelist allowed functions
ALLOWED_FUNCTIONS = {"run_test", "analyze_coverage"}

async def execute_function(func_name: str, *args):
    if func_name not in ALLOWED_FUNCTIONS:
        raise ValueError(f"Function {func_name} not allowed")
    func = globals()[func_name]
    return await alcall(func, *args)

# ❌ BAD: Unrestricted function access
async def execute_function(func_name: str, *args):
    func = globals()[func_name]  # Can execute ANY function!
    return await alcall(func, *args)

# ✅ GOOD: Validate paths before operations
from pathlib import Path

def read_file(file_path: str):
    path = Path(file_path).resolve()
    if not path.is_relative_to(PROJECT_ROOT):
        raise ValueError("Path traversal detected")
    return path.read_text()

# ✅ GOOD: Implement rate limiting
from lionagi_qe.core.hooks import enable_cost_alerts

enable_cost_alerts(
    cost_threshold=10.0,  # Alert if cost > $10/hour
    rate_limit=100  # Max 100 calls/minute
)

• Risk: Executes arbitrary test commands
• Mitigation: Always validate test file paths, disable shell=True
• Status: Fixed in v1.0.0

• Risk: Operates on filesystem
• Mitigation: Implement path validation, restrict to project directory
• Status: Path validation added in v1.0.0

• Risk: Tracks sensitive cost/usage data
• Mitigation: Encrypt data at rest, restrict access to authorized users
• Status: Encryption recommended (not yet implemented)

• Risk: Potential prompt injection attacks
• Mitigation: Sanitize inputs, validate outputs, implement rate limiting
• Status: Basic validation in place

1. Command Injection (CVE-TBD, CVSS 9.8)

   • Component: test_executor.py
   • Issue: shell=True allowed arbitrary command execution
   • Fix: Removed shell=True, switched to list-based arguments
   • Affected Versions: 0.1.x
   • Fixed in: 1.0.0

2. Arbitrary Code Execution (CVE-TBD, CVSS 9.1)

   • Component: alcall wrapper
   • Issue: Unrestricted function access via globals()
   • Fix: Added function whitelist validation
   • Affected Versions: 0.1.x
   • Fixed in: 1.0.0

3. Insecure Deserialization (CVE-TBD, CVSS 8.8)

   • Component: hooks.py
   • Issue: pickle.loads() on untrusted data
   • Fix: Replaced with JSON deserialization
   • Affected Versions: 0.1.x
   • Fixed in: 1.0.0

4. Path Traversal (CVE-TBD, CVSS 7.5)
   • Component: code_analyzer.py
   • Issue: Missing path validation
   • Fix: Added path validation and sandboxing
   • Affected Versions: 0.1.x
   • Fixed in: 1.0.0

See CHANGELOG.md for complete release notes.

• All code changes require security review
• Automated SAST scanning in CI/CD
• Dependency vulnerability scanning

• Security-focused test cases
• Penetration testing for major releases
• Fuzzing for input validation

• Regular dependency updates
• Automated vulnerability scanning
• Minimal dependency footprint

1. Day 0: Vulnerability reported privately
2. Day 2: Initial acknowledgment sent
3. Day 7: Assessment complete, fix timeline established
4. Day 14-30: Patch developed and tested
5. Day 30: Patch released to supported versions
6. Day 37: Public disclosure (7 days after patch)

• 2025-11-05: Internal security audit (qe-security-scanner agent)
  • 3 critical vulnerabilities found and fixed
  • 3 high-priority issues addressed
  • Security score improved from 68/100 to 95/100

• OWASP Top 10
• CWE Top 25
• Python Security Best Practices

• Security Email: security@lionagi-qe-fleet.example.com
• General Issues: GitHub Issues
• Documentation: Security Best Practices

Last Updated: 2025-11-05 Version: 1.0.0