This is a deep technical map of Turso's Change Data Capture implementation, tracking INSERT/UPDATE/DELETE operations at the bytecode level. It's valuable if you're working on sync engines, debugging CDC behavior, or need to understand how changes flow from SQL statements through the translate layer into the turso_cdc table and eventually to remote replicas. The architecture section shows exactly where CDC hooks into the execution pipeline, and the version detection logic (v1 vs v2 schemas, transaction grouping, COMMIT records) is spelled out in detail. Worth reading if you're touching anything in core/translate or sync/engine, or if you've ever wondered why CDC and MVCC can't run simultaneously.
- skills.sh installs
- 153
- GitHub stars
- 25.2K
- last updated
- Aug 21, 2026
Install
$ npx -y skills add tursodatabase/turso --skill cdc --agent claude-codeInstalls into .claude/skills of the current project.
Files
CDC (Change Data Capture) - Internal Feature Map
Overview
CDC tracks INSERT/UPDATE/DELETE changes on database tables by writing change records into a
dedicated CDC table (turso_cdc by default). It is per-connection, enabled via PRAGMA, and
operates at the bytecode generation (translate) layer. The sync engine consumes CDC records
to push local changes to the remote.
Architecture Diagram
Sponsors
Details
- Repository
- github.com/tursodatabase/turso
- Categories
- Code Review & Quality, Databases
- First seen
- Jun 3, 2026
