If you discover a security vulnerability in the SCMS Starter Kit, please report it responsibly:
π§ Email: manny@aialchemist.net
Please include:
- Description of the vulnerability
- Steps to reproduce
- Potential impact assessment
- Suggested fix (if any)
- Open a public GitHub issue for security vulnerabilities
- Disclose the vulnerability publicly before we've addressed it
- Initial Response: Within 48 hours
- Assessment: Within 1 week
- Fix Timeline: Depends on severity (critical issues prioritized)
- Disclosure: After fix is deployed and users have time to update
SCMS is designed with security and privacy as core principles:
β No Shared Secrets
- Users supply their own AI API keys
- No central authentication required
- No shared credentials to compromise
β Local-First Data Storage
- All user data stored on local filesystem
- No cloud storage or databases
- No telemetry or tracking
β Transparent & Auditable
- Open source codebase
- No obfuscated code
- All dependencies visible
β Zero Trust Model
- No network dependencies for core functionality
- No external API calls from scripts
- Dashboard reads local files only
β No personal information
β No usage analytics
β No telemetry data
β No AI prompts or responses
β No API keys or credentials
β No user behavior tracking
Protected via .gitignore:
- API keys (
.env.local) - User tracking data (
economics-dashboard-data.json) - Session histories (
checkpoints/,scms-sessions/) - Custom configurations (
WORKSPACE_RULES.custom.md) - Backup files (
*.backup.json)
Users are responsible for:
- Securing their own AI API keys
- Managing file permissions on their system
- Not committing sensitive data to their own repositories
- Following security best practices in their IDE
β DO:
- Store API keys in
.env.local(excluded from git) - Use environment variables for sensitive data
- Rotate keys regularly
- Use separate keys for development/production
β DON'T:
- Hardcode API keys in any files
- Commit
.env.localto version control - Share API keys in documentation or comments
- Use production keys for testing
When creating custom workspace rules:
- Use
WORKSPACE_RULES.custom.md(excluded from git) - Avoid including sensitive project details
- Don't store credentials in rules files
- Review before committing to ensure no secrets
The economics dashboard creates economics-dashboard-data.json:
- This file is excluded from git
- Contains token usage and session data
- Safe to share if you want to show SCMS effectiveness
- Delete anytime to reset tracking
If using checkpoint monitoring:
checkpoints/directory is excluded from git- Contains AI conversation summaries
- May include project-specific context
- Treat as private unless you explicitly want to share
Enhanced .gitignore patterns protect:
- API keys and certificates (
.pem,.key,.cert) - Environment files (
.env*.local) - Database files (
.db,.sqlite) - Secrets and credentials (
secrets.json,credentials.json) - Temporary files and logs
Dependencies are minimal:
got- HTTP client (for checkpoint monitoring only)cheerio- HTML parsing (for dashboard only)- Standard Node.js modules
Security monitoring:
- Enable Dependabot alerts on GitHub
- Review dependency updates carefully
- Keep dependencies up to date
Safe practices:
- No
eval()or dynamic code execution - No shell command injection vectors
- File system access is read-only except for user data directories
- No external network calls except explicit checkpoint monitoring
When a security vulnerability is reported and confirmed:
- Acknowledgment: We'll confirm receipt within 48 hours
- Assessment: We'll evaluate severity and impact
- Fix Development: We'll develop and test a fix
- Notification: We'll notify the reporter before public disclosure
- Release: We'll release a patched version
- Disclosure: We'll publish a security advisory after users have time to update
Severity Levels:
- Critical: Immediate attention, patch within 24-48 hours
- High: Patch within 1 week
- Medium: Patch within 2-4 weeks
- Low: Patch in next regular release
SCMS is a tool for improving AI-assisted development workflows. Users should:
β Ethical Use:
- Respect intellectual property when storing patterns
- Follow your organization's data policies
- Don't store proprietary code in public repositories
- Comply with your AI provider's terms of service
β Privacy:
- Don't store personally identifiable information in SCMS files
- Don't commit client or customer data
- Treat your workspace rules as private documentation
β Legal Compliance:
- Comply with applicable laws and regulations
- Respect software licenses when documenting patterns
- Follow your employment agreements regarding code ownership
Stay informed:
- Watch this repository for security advisories
- Check the CHANGELOG for security-related updates
- Subscribe to release notifications
Apply updates promptly:
cd scms-starter-kit
git pull origin master
npm installIf you have security questions that aren't vulnerabilities:
- Open a GitHub Discussion
- Email: manny@aialchemist.net
- Check existing documentation
For vulnerabilities, always use private email reporting.
Last Updated: January 30, 2026
Version: 1.0