Back to Home
Privacy & Data Stewardship

Privacy Policy & Data Principles

ContribLens is built from the ground up on privacy-preserving, evidence-first principles. We do not monetize personal data or retain repository code.

1. Zero Raw Secret Storage

When scanning repositories, any matching API tokens, AWS keys, GitHub PATs, private keys, or passwords are automatically and irreversibly redacted using in-memory regex masking (ghp_****). Raw secrets never touch the database, logs, or AI prompt contexts.

2. Zero Code Execution

ContribLens never runs arbitrary code, executes scripts, or executes package manager installers (npm install, pip install) against audited repositories. All analysis is purely static, AST/manifest parsed, and read-only.

3. Safe AI Contract & Minimal Bundles

We send the AI adapter minimal, redacted evidence bundles rather than whole-repository dumps. Gemma and Gemini are constrained to explain validated facts only and cannot alter findings, invent CVEs, or transmit repository contents outside our secure adapter boundary.

4. User Rights, Export & Data Portability

You own your scan history and generated issue drafts. From your Profile settings, you can export all stored records as a formatted JSON bundle or permanently purge your repository activity and scan history at any time.

© 2026 ContribLens. Built for open source safety.