Documentation

Brume produces GDPR-aligned copies of your PostgreSQL data for debugging, testing and demos — without ever shipping real identities.

This documentation is organized as a journey: discover, understand, look things up, operate in production.

Start here

New to Brume? Begin with these two pages.

• Getting started — install Brume on Debian, Fedora or macOS and check the install.
• Your first pseudonymization — write a minimal brume.yml, run brume plan then brume execute, end-to-end in five minutes.

Concepts

The mental model behind the tool — read these once and the rest of the docs make sense.

• How Brume works — extraction, transformation, write phases. Determinism. Secrets. FK traversal.
• Pseudonymization strategies — FAKE, MASK, HASH, NULLIFY, FPE_ID, FPE_UUID, DATE_SHIFT, KEEP. When to pick which, and how they behave on UNIQUE columns.
• Semantic types — EMAIL, PHONE, IBAN… how FAKE and MASK use them.

Reference

The exhaustive doc for every flag, every key, every variable.

• brume.yml — full schema of the configuration file.
• CLI commands — every subcommand, every flag, exit codes.
• .env.brume variables — secrets, connection strings, runtime options.

Use it well

Production-grade usage, compliance, and troubleshooting.

• Recipes — end-to-end task-oriented walkthroughs (refresh staging, multi-tenant, JSONB, CI/CD…).
• GDPR & compliance — legal position, GDPR articles covered, DPO checklist.
• k-Anonymity audit — measure residual re-identification risk.
• Operations & troubleshooting — performance, secret management, common errors.

---

Prefer step-by-step guides? See Guides. Looking for the roadmap? See Roadmap. Need help? See Support.