How does Metaxy handle data versioning?

It tracks sample versions and allows manual version bumping via 'code_version' in FieldSpec to trigger downstream recomputation automatically.

Does Metaxy support AI assistants via MCP?

Yes, Metaxy includes a built-in Model Context Protocol (MCP) server that can be started using the 'mx mcp' command.

Metaxy is a metadata layer for managing feature versions, dependencies, and data lineage in complex multi-modal Data and ML pipelines.

Can I visualize my feature dependencies?

Yes, you can use the 'mx graph render' CLI command to visualize the feature dependency graph directly in your terminal.

Which databases does Metaxy support?

It supports multiple storage backends including DuckDB, ClickHouse, BigQuery, LanceDB, and Delta Lake.

Metaxy Metadata Manager

Name: Metaxy Metadata Manager
Author: anam-org

byanam-org

•

Data Science & ML

Manages and tracks feature versions, dependencies, and data lineage for multi-modal data and machine learning pipelines.

About

Metaxy provides a specialized metadata layer for complex computational graphs, enabling developers to define features with explicit field-level dependencies and automated version tracking. It simplifies the management of incremental pipelines by propagating changes through a feature graph and supporting diverse storage backends like DuckDB, ClickHouse, and Delta Lake. By using this skill, developers can automate recomputation triggers via code versioning, visualize data lineage through a dedicated CLI, and maintain strict metadata standards across multi-modal AI projects.

Key Features

Multi-backend support including DuckDB, ClickHouse, and Delta Lake
Automated dependency graph management and visualization
20 GitHub stars
Integrated CLI for migrations, status checks, and MCP server
Field-level data lineage and incremental processing logic
Versioned feature definitions with BaseFeature and FieldSpec

Use Cases

Building incremental multi-modal data pipelines for ML models
Tracking and propagating version changes across large data graphs
Managing complex feature dependencies in production AI systems

About

Key Features

Multi-backend support including DuckDB, ClickHouse, and Delta Lake
Automated dependency graph management and visualization
20 GitHub stars
Integrated CLI for migrations, status checks, and MCP server
Field-level data lineage and incremental processing logic
Versioned feature definitions with BaseFeature and FieldSpec

Use Cases

Building incremental multi-modal data pipelines for ML models
Tracking and propagating version changes across large data graphs
Managing complex feature dependencies in production AI systems