We release patches for security issues in the following versions:
| Version | Supported |
|---|---|
| 1.x.x | ✅ |
| < 1.0 | ❌ |
Please do not report security issues through public GitHub issues.
If you discover a security issue, please report it by emailing security@syrin.ai.
Include the following in your report:
- Description of the issue
- Steps to reproduce (if applicable)
- Potential impact
- Any suggested fixes (optional)
-
Acknowledgment: We will acknowledge receipt of your report within 48 hours.
-
Assessment: We will investigate and determine the severity and impact.
-
Updates: We will keep you informed of our progress.
-
Resolution: Once resolved, we will notify you and discuss public disclosure timing.
-
Credit: We are happy to credit reporters in our release notes (unless you prefer to remain anonymous).
This security policy applies to:
- The
@syrin/clinpm package - The Syrin CLI tool and its dependencies
- The official documentation at docs.syrin.dev
- Third-party MCP servers connected via Syrin
- User-configured LLM provider credentials (these are your responsibility to secure)
- Issues in dependencies (please report these to the respective maintainers)
-
Protect your API keys: Use environment variables or the global
.envfile (~/.syrin/.env), never commit credentials to version control. -
Review tool contracts: Before enabling
--execmode, understand what tools will be executed. -
Use sandboxed testing: The
syrin testcommand runs tools in isolated environments with resource limits. -
Keep Syrin updated: Run
syrin updateregularly to get security patches.
- Execution disabled by default:
syrin devrequires explicit--execflag - Sandboxed test execution: Resource limits and process isolation
- Contract validation: Static analysis catches issues before runtime
- Event audit trail: All tool executions are logged to
.syrin/events/