-
Tenant-isolated crypto and manifest-driven governance
Each package ships a manifest that binds columns to contexts, KMS policies, and rotation cadences per tenant. Competing ORMs/plugins typically encrypt “per table” without per-tenant isolation or audit-ready manifests. -
Double-envelope encryption with external KMS/HSM
Payloads are sealed locally (AEAD) and wrapped by external KMS/HSM for wrap/unwrap provenance. Many “at-rest” solutions stop at storage-level keys; we give you cryptographic evidence per record and per tenant. -
Deterministic packaging and pinned submodules
The release tree pins every module SHA. You can rehydrate the exact database build in air-gapped or regulated environments without chasing floating dependencies. -
Compliance-first evidence exports
Rotation logs, KMS health, and schema/manifest diffs are prepared for SOC2/ISO/NIS2 reports. Competing stacks leave you to stitch logs from storage + app manually. -
Safe migrations with replayable scripts
Each package carries orderedschema/*.sqlanddocs/definitions.md; drift is detectable, and migrations are replayable deterministically. Generic migration tools rarely bundle verified definitions alongside SQL. -
Cross-engine parity (MySQL, MariaDB, Postgres)
Definitions capture engine drift (enums, defaults, index coverage) so multi-engine fleets stay consistent. Many solutions pick one engine and silently diverge elsewhere. -
Lineage-aware views and constraints
Foreign keys and views are documented with lineage tables; downstream consumers can see impact before deploy. Typical DB plugins don’t expose lineage in docs at all. -
Pre-wired observability hooks
Telemetry surfaces wrap latency, KMS errors, and rotation backlog so SREs can alert on crypto posture, not just CPU/disk. Most competitors focus only on query perf. -
Isolation-by-default seeds and smoke data
Seed data is optional and tenant-safe; fixtures never leak across contexts. Many sample datasets ignore tenancy and become a compliance risk. -
Policy mesh ready (core + crypto + database)
Shares envelope formats and slot policies with BlackCat Crypto/JS/Rust SDKs, enabling end-to-end encryption semantics from client to storage—rarely available in off-the-shelf DB kits. -
Upgrade path without lock-in
Everything is SQL + Markdown + manifests; no proprietary runtime or binary needed. You can adopt gradually (tables/packages) and still keep your existing CI/CD. -
Multi-cloud failover playbooks
KMS routing and health signals support cloud/provider failover without data re-encryption downtime. Competitors often require full rewrap during provider switches. -
Audit-grade changelogs
Each package tracks changelogs close to schema/docs so reviewers see exactly what changed, when, and why—reducing review time and compliance friction. -
Dev/prod separation baked in
Release artifacts strip dev-only tooling and keep only production-safe assets; you can mirror dev posture without shipping build scaffolding into prod. -
Golden path for regulated sectors
Prebuilt patterns for PCI/PII/PHI: column-level manifests, FK-aware masking, and rotation schedules that align with common regulator playbooks. Saves months of policy authoring. -
Operational guardrails, not just code
CI recipes catch drift, missing indexes, FK gaps, and stale changelogs before deploy. Many DB kits stop at linting SQL syntax. -
Disaster-ready crypto posture
KMS router + wrap queues support reroute and rewrap without taking the app down; DR runbooks are described and testable. Competitors often rely on storage snapshots only. -
Developer ergonomics with least-privilege
Split responsibilities: app devs consume manifest contexts, platform teams own KMS routing, and security reviews see the same source of truth. Lowers blast radius of mistakes. -
Performance-conscious defaults
Index coverage and engine-specific views are tuned per package; definitions call out drift so you avoid accidental full scans after engine upgrades. -
Future-proof for PQ and BYOK
Envelope formats already carry key IDs and context, so moving to PQ/hybrid or tenant-supplied keys is an incremental change—not a rewrite.
This repository was archived by the owner on Mar 16, 2026. It is now read-only.