Security
Last updated October 2, 2026
Relayza treats workspace isolation, credential protection, bounded provider actions, operational monitoring, and recoverability as launch requirements. This page describes the controls used during the controlled beta and how to report a suspected vulnerability.
Current security practices
- database Row Level Security and workspace-membership authorization;
- controlled RPC boundaries for sensitive content and platform operations;
- server-side storage of provider credentials and service credentials;
- single-use, expiring OAuth state bound to user, workspace, and provider;
- rate limits, quotas, retry budgets, and platform kill switches;
- Content Security Policy and private-route noindex protections;
- structured redacted operational telemetry and audited operator actions;
- owned backup exports plus clean-room restore drills.
Good-faith vulnerability reporting
We welcome clear reports that help protect Relayza and its users. Please avoid privacy violations, social engineering, denial-of-service activity, destructive testing, persistence, credential theft, or accessing data beyond what is necessary to demonstrate a problem.
Do not test third-party systems such as Meta, Supabase, Vercel, Resend, Google, GitHub, or Cloudflare under this guidance; their own policies apply.
Report securely
Describe the affected Relayza route or feature, reproduction steps, impact, and any safe proof of concept. Redact credentials and personal data. Relayza does not currently operate a paid bug-bounty program and does not promise payment for reports.