| Version | Supported |
|---|---|
| 9.0.x | ✅ Yes |
| 8.2.x | ✅ Security fixes |
| 8.1.x | ❌ No |
| < 8.0 | ❌ No |
Please do NOT open public GitHub issues for security vulnerabilities.
Instead, use GitHub's private vulnerability reporting:
- Go to the repository's Security tab
- Click "Report a vulnerability"
- Fill out the form with details
- Submit privately
Link: https://github.com/LgrappaG/Workflows-Agents/security/advisories
Email security concerns to: [security@example.com]
Include:
- Description of the vulnerability
- Steps to reproduce (specific commands or scenarios)
- Potential impact (what could an attacker do?)
- Suggested fix (if available)
- Your contact information
We care about reports involving:
- Code Injection: Unsanitized inputs in skill definitions or workflows
- XSS/Script Injection: Malicious code execution in documentation
- Dependency Vulnerabilities: Compromised or outdated packages
- Information Disclosure: Accidental exposure of sensitive data
- Authentication/Authorization flaws: Improper access controls
- Validation Bypass: Circumventing safety checks
We do NOT classify these as vulnerabilities:
⚠️ Social engineering or phishing attempts⚠️ Physical security issues⚠️ Feature requests disguised as bugs⚠️ Third-party service issues (e.g., GitHub, npm)
| Stage | Timeline |
|---|---|
| Initial Response | Within 24 hours |
| Investigation | Within 5 business days |
| Patch Release | Within 14 days (critical) or next regular release |
| Public Disclosure | After patch is released or 90 days, whichever is sooner |
- Acknowledgment: We'll confirm receipt of your report
- Investigation: We'll verify and assess the issue
- Coordination: We'll discuss timeline and fix with you
- Patch: We'll develop and test a fix
- Release: We'll release a patch update (or minor version)
- Public Notice: We'll publish a security advisory with credit to the reporter (unless you prefer anonymity)
When contributing skills and workflows, please ensure:
# ❌ Bad - allows arbitrary input
username = user_input
# ✅ Good - validates input
if not isinstance(username, str) or len(username) > 255:
raise ValueError("Invalid username")# ❌ Bad
api_key: "sk-1234567890abcdef"
# ✅ Good - reference environment variables
api_key: ${OPENAI_API_KEY}# If processing Unity project files:
import os
if os.path.isabs(path): # Prevent directory traversal
raise ValueError("Path must be relative")Review these common vulnerabilities in game dev contexts:
- Injection: SQL, command, template injection
- Broken Authentication: Weak credential handling
- Exposed Data: Unencrypted sensitive information
- Broken Access Control: Bypass authorization checks
- Configuration Issues: Default credentials, verbose errors
- Vulnerable Dependencies: Outdated packages
- Authentication Failure: Poor session management
- Data Integrity Issues: Malicious file uploads
- Insufficient Logging: No audit trails
- SSRF: Server-side request forgery
- Monthly: Dependency scanning with
npm auditandpip audit - Quarterly: Manual code review by security-focused contributors
- Yearly: Third-party security assessment (if resources allow)
We keep dependencies minimal and regularly updated:
# Check for vulnerabilities
npm audit
pip audit
# Update safely
npm update
pip install --upgrade -r requirements.txt.agents respects:
- GDPR: No personal data collection
- OWASP Top 10: Security best practices
- CWE: Common Weakness Enumeration standards
- CVE: Tracking known vulnerabilities
For security questions (non-vulnerability):
- Open a GitHub Discussion in the Security category
- Email:
security@example.com
Thank you for helping keep .agents secure! 🔒