Security at RodGrid

Protecting the work you trust us with.

RodGrid is built to protect account, workspace, and interview data without hiding behind vague compliance language. This page explains the controls in place today, how we respond when something goes wrong, and where our assurance program is still growing.

Protected in transit
Browser and service traffic uses encrypted HTTPS and secure WebSocket connections.
Private by default
Workspace, Grid, interview, and scorecard access is enforced by authenticated roles and scoped links.
Card data stays with Paddle
RodGrid does not receive or store raw payment card numbers.

Security controls

How RodGrid protects your data

These controls describe the current product and operating posture. They are not a claim of certification.

Identity and access

OAuth and passwordless sign-in, optional user two-factor authentication, session expiry and revocation, role-based access, and single-use hashed invitation tokens protect account and workspace access.

Application security

Authorization checks, request limits, input validation, security headers, signed billing webhooks, and focused regression tests protect the application boundary.

Data minimization

RodGrid avoids raw card storage, hashes security tokens and audit identifiers where practical, and limits collected information to operating and protecting the service.

Service providers

Cloudflare, Paddle, OpenAI, Resend, GitHub, and Google support defined parts of the service. Data flows and subprocessors are disclosed in the Privacy Policy.

Retention and deletion

Account and Grid deletion, backup windows, export options, and configurable interview retention help reduce how long information remains in the service.

Review and monitoring

Automated tests, secret scanning, dependency vulnerability scanning, operational logging, health checks, and code-level security reviews are used to detect regressions and investigate issues.

Incident response

What happens when something goes wrong

A lightweight response process gives us a clear sequence to follow instead of improvising during a security event.

  1. Receive and assess

    We record the report, preserve relevant evidence, determine severity, and identify affected systems and people.

  2. Contain and investigate

    We revoke access, rotate credentials, disable affected paths, or take other proportionate steps while determining the cause and scope.

  3. Recover and communicate

    We fix and verify the issue, monitor for recurrence, and notify affected people or authorities when required by applicable law.

Current assurance

Clear about what we have—and what we do not

Trust should come from specific, verifiable statements. We do not present future work as a completed control.

  • RodGrid performs internal, code-level security reviews of security-sensitive application paths.
  • Automated security regressions and secret scanning are part of the delivery checks.
  • Customer content stored in RodGrid's Cloudflare D1, R2, and Durable Object storage is encrypted at rest by the platform.
  • RodGrid runs its own internal dynamic application security testing—automated scanning (OWASP ZAP) plus manual access-control and business-logic checks against a staging build—which is our own work, not a substitute for an independent third-party assessment.
  • RodGrid does not currently have a SOC 2 report.
  • RodGrid has not yet completed an independent third-party penetration test.

Responsible disclosure

Report a vulnerability

If you believe you found a security weakness in RodGrid, please send it privately to our security inbox. We appreciate clear, good-faith reports and will review them as quickly as practical.

security@rodgrid.io

Helpful details to include

  • A concise description of the issue and its potential impact.
  • Steps to reproduce, including the affected URL or feature.
  • Any screenshots, logs, or proof that do not expose another person's data.
  • A safe way to contact you with follow-up questions.

Please do not access other users' information, disrupt the service, use social engineering, or publicly disclose an unresolved issue.