What it is

Bytebase is an open-source database governance platform that acts as a single control plane between human users, AI agents, and databases. It governs change management, access control, and compliance so that every database change and query can be reviewed, controlled, and recorded. The project sits in the Infrastructure and Operations database ecosystem, with topics that include CI/CD, DevOps, DevSecOps, GitOps, data governance, data masking, data security, database access, Flyway, Liquibase, and MongoDB.

The concrete problem it addresses is the fragmentation of database operations across migration scripts, SQL clients, and ticketing systems. Bytebase unifies those workflows in one platform, supporting PostgreSQL, MySQL, SQL Server, Oracle, MongoDB, Redis, MariaDB, TiDB, Snowflake, ClickHouse, Spanner, and OceanBase. It also lists integrations spanning infrastructure as code, AI, identity providers, collaboration, ITSM, log streaming, and secret managers.

Key capabilities

• Change management through a web console lets users request, review, deploy, and rollback database changes.
• GitHub and GitLab integration supports database-as-code workflows through GitOps.
• SQL Review applies 200+ lint rules to enforce SQL standards and best practices.
• Fine-grained RBAC provides project and workspace-level roles and permissions.
• Just-in-time access grants time-boxed database access with automatic revocation.
• Dynamic data masking applies column-level masking based on user role at query time.
• Audit logging records database activities, while Terraform Provider and API manage policies as code.

Who uses it and how

• Development teams implement database schema version control, automate database deployments through CI/CD pipelines, and collaborate on changes with review workflows.
• DBAs centralize database management across environments, enforce organization-wide SQL standards and policies, and monitor and audit database activities.
• Security teams control data access with column-level permissions, implement data masking for sensitive information, and maintain compliance with audit trails.
• AI agents and IDEs connect through the MCP Server, while the SQL Editor offers Text-to-SQL and Page Agent assistance for query generation and workflow execution.

Getting started

Typical self-hosted installation uses Docker with the bytebase/bytebase:latest image, publishing port 8080 and mounting ~/.bytebase/data, or Kubernetes with the bytebase/bytebase Helm chart; users then visit http://localhost:8080 and follow the setup wizard.

When to use it — and when not to

Choose Bytebase when teams need one platform for database change review, access governance, and audit trails instead of stitching together migration scripts, SQL clients, and ticketing systems. The README lists comparisons with Liquibase, Flyway, and Jira, so it is relevant where those workflows are fragmented. Self-hosting requires operating the Docker or Helm deployment, the contributor setup shows a PostgreSQL dependency through PG_URL, and the provided metadata lists zero contributors and 167 open issues.

docker run --init \
  --name bytebase \
  --publish 8080:8080 \
  --volume ~/.bytebase/data:/var/opt/bytebase \
  bytebase/bytebase:latest

helm install bytebase bytebase/bytebase

# Setup a postgres database with user bbdev and database bbdev
export PG_URL=postgresql://bbdev@localhost/bbdev

# Start backend
alias r='go build -ldflags "-w -s" -p=16 -o ./bytebase-build/bytebase ./backend/bin/server/main.go && ./bytebase-build/bytebase --port 8080 --data . --debug'

# Start frontend
alias y="pnpm --dir frontend i && pnpm --dir frontend dev"