Skip to content

Security: Blake-Daniel-Inv/SnowCrew-AI

Security

SECURITY.md

Security Policy

Reporting a Vulnerability

Please do not open a public issue for security vulnerabilities.

Report privately using GitHub's Private Vulnerability Reporting: go to the Security tab → Report a vulnerability. This keeps the report confidential until a fix is available.

Reports are typically acknowledged within a few days, and you'll be kept updated on remediation progress.

Supported Versions

This project is pre-1.0 and developed on a rolling basis. Only the latest main receives security fixes.

Version Supported
main (latest)
older commits

Scope & Threat Model

SnowCrew AI is designed for local / single-tenant deployment. A few things worth knowing when assessing a report:

  • Credentials at rest. Third-party tokens (GitHub OAuth, Snowflake PAT/JWT) are envelope-encrypted before storage using the key in CREDENTIALS_MASTER_KEY. Protecting that key — and your .env* files — is the operator's responsibility. Losing the key means losing stored credentials; exposing it compromises them.
  • Secrets stay server-side. Readiness/health endpoints return variable names, never values.
  • Generally out of scope: issues that require an already-compromised host, a leaked CREDENTIALS_MASTER_KEY, or real secrets committed to your own fork.

If you're unsure whether something is in scope, report it privately and we'll triage.

There aren't any published security advisories