braidplane-core

README

Braidplane Community core

Braidplane is a declarative, contract-based data integration platform.

Define ETL pipelines in Git, specify data contracts between steps, and inspect the exact version and results of each run. Braidplane is open source and built for developers and data teams who move and combine data across databases, APIs, CRMs and file stores. For example, join order records with customer data and write the result to S3. Work through the web interface, CLI or API.

How it works · Run locally · Pipeline examples · Connector capabilities

Run lineage with synthetic data: settlement file + payment ledger → comparison rows and totals. Inspect the executed pipeline version, step status, timing and row counts. Click the image for a full-size view.

Explore pipeline examples · How runs work

What you can do

• Connect PostgreSQL, MySQL, MongoDB, HTTP APIs, HubSpot, S3 and other available systems.
• Review pipeline changes in Git, with credentials stored separately in configured connections.
• Use different connections per environment. Bind a definition's logical connection names to a project's saved data sources.
• Validate the definition, deploy an immutable version, then inspect run status, step timelines and failure diagnostics.
• Assess changes with Proof. Inspect claims and evidence for an exact Git change and target before deciding whether to apply it.

See the connector guide for available operations, connection settings and examples.

Community edition

Community includes the backend, web interface, CLI, API schema, core migrations, Proof, authentication, authorization and run diagnostics. There are no pipeline-count, usage or trial restrictions. Configure runtime resource limits through CONFIG.md.

Run locally

From a checkout of this repository, install Bash, Git, Python 3, OpenSSL, Go, Task and Docker with Compose. Start Docker, then run:

task dev

The first run downloads the pinned toolchains and dependencies, builds the application, starts PostgreSQL and Mailpit, and applies core migrations. Open http://127.0.0.1:3002, create an account and find its verification email at http://127.0.0.1:38025. Provider credentials are only needed for integrations that use them.

Ctrl-C stops the application processes; the database, Mailpit and local keys remain for the next run. Keep the keys with the database: encrypted data needs the original keys. See local setup for ports, stopping containers and running multiple instances. To grant administrator access, follow instance administrators.

Build your first pipeline with the PostgreSQL to S3 walkthrough. It includes sample data, a pipeline definition, expected results and cleanup commands. You will need a database, an S3 bucket and a Git repository.

Common questions

Is AI used to develop Core?

Yes. AI tools are used during development, testing and review. The maintainer decides which changes are accepted and remains responsible for the project.

Can I use AI tools when contributing?

Yes. You remain responsible for the changes you submit. See AI-assisted contributions for disclosure, testing and reporting requirements.

Help and participation

Use GitHub Issues for bugs and proposals, support guidance for maintenance and conduct, and SECURITY.md for private vulnerability reports. Read CHANGELOG.md for changes and the contribution guide before submitting work.

Develop

Use build and validation for toolchain, test, generation and container instructions. Contributing and quality gates describe checks for changes; engineering documentation, configuration and package boundaries cover runtime settings and code ownership.

License

First-party code and technical documentation use Apache-2.0. Third-party files retain their own licenses; see NOTICE. See the release rights inventory for third-party components and notice requirements.