Privacy Policy

This Privacy Policy details the data safety practices and architecture of ScrubBeforeAI. Because we value developer privacy and absolute secure boundaries above all else, our data design is simple: we do not collect, transmit, store, or see your data.

1. Client-Side Text Processing

All operations related to log scanning, regex matching, placeholder replacement, and prompt un-redacting occur entirely within your web browser sandbox. No backend API servers, databases, or cloud processing layers are used.

Because we use the native browser JavaScript RegExp engine, the text you paste never leaves your local computer.

2. Local Session Map Storage

To restore placeholder tags back into original secrets (the "Un-Redact" process), the application holds a temporary key-value lookup map.

This lookup mapping is stored strictly using browser-level sessionStorage and in-memory React application state.

• Tab Scoped: Session storage is isolated per browser tab. One tab cannot inspect another's key map.
• Auto-Purge: The reversal map is automatically cleared and destroyed when you close the browser tab.
• No Offline Persistence: Unlike localStorage, session data is not permanently written to your disk and is lost upon session termination.

3. Zero Telemetry & Tracking

We do not use any analytical packages, tracking scripts, or event loggers (such as Google Analytics or Amplitude).

We do not track what parameters you redact, how often you use the tool, or your IP address. The site is a fully static application that functions offline once loaded.

4. Compliance and Data Governance

By sanitizing raw codebases and terminal logs locally, developers ensure compliance with GDPR, HIPAA, PCI-DSS, and corporate security guidelines. ScrubBeforeAI serves as a client-side boundary protecting your team from inadvertent data leakages into third-party LLM training logs.