Import + Amend via Schema in API, CLI, and MCP
· One min read
The Import + Amend via Schema flow is now available across all automation interfaces: REST API, CLI, and MCP.
You can now provide a schema plus existing input data, amend rows using schema rules, and get back the full resulting dataset in your requested output format.
What Was Added
- REST:
POST /v1/generate/amend - CLI:
anywaydata amend --schema-file ... --data-file ... --input-format ... - MCP:
amend_data_from_spec
All three share the same core behavior.
Behavior
rowCountdefaults to imported row count.- If
rowCountis smaller, only firstNrows are amended. - Output always returns the full resulting dataset after amendment.
streamis accepted for compatibility but ignored for amend operations (buffered mode only).
Cross-Format Support
The flow supports import and export across formats, including:
- CSV input -> tab-delimited (
dsv) output - tab-delimited (
dsv) input -> CSV output
Why This Matters
- You can automate the same amend workflow previously available in UI.
- You can update existing datasets without rebuilding them from scratch.
- You can keep the same schema-driven logic across UI, REST, CLI, and MCP.
