---
title: "data-engine-kit"
description: "End-to-end data engineering toolkit for BigQuery, dbt, Cube Core, and Airflow workflows."
canonical: https://agentpluginsdirectory.com/plugins/data-engine-kit
last-updated: 2026-09-22
---

# data-engine-kit
End-to-end data engineering toolkit for BigQuery, dbt, Cube Core, and Airflow workflows.
- Slug: data-engine-kit
- Publisher: juanpabosu
- Repository: https://github.com/juanpabosu/data-plugins
- Manifest: plugins/data-engine-kit/plugin.json
- Version: 0.1.0
- Category (editorial): databases
- Skills: 4 (derived-metrics-cube, generate-cube, pre-agg-cube, validate-cube)
- MCP servers: 2 (bigquery, cube)
- Stars: 0
- Repository created: 2026-08-07
- Repository last pushed: 2026-08-07
- Publisher type: User
- Listing: https://agentpluginsdirectory.com/plugins/data-engine-kit
- Schema: https://agent-plugins.org/schemas/1.0.0/plugin.schema.json

## What data-engine-kit does, in the publisher's words

End-to-end data engineering toolkit for BigQuery, dbt, Cube Core, and Airflow workflows.

From the project README, punctuation lightly normalized. Full text: https://raw.githubusercontent.com/juanpabosu/data-plugins/HEAD/plugins/data-engine-kit/README.md

## Skills

- derived-metrics-cube: Adds derived metric measures (ratio, average, filtered count) to an existing cube YAML file and updates the view includes list. Works on any cube already generated by generate-cube.
- generate-cube: Generates Cube.dev cube + view YAML files for any core_* BigQuery dbt model. Runs generate_cube.py for deterministic column mapping, then iterates with the developer to resolve INT64 ambiguities, PK, derived metrics, and pre-aggregations before writing final files.
- pre-agg-cube: Adds pre-aggregation blocks to an existing cube YAML file based on table size, partition strategy, and cluster keys from INFORMATION_SCHEMA and the dbt manifest. Run after generate-cube when the model is large or frequently queried.
- validate-cube: Validates the local Cube semantic layer by starting the Docker instance, polling until ready, confirming all cube YAML files load via the /meta API, then stopping. Returns a short pass/fail summary. Use after generating or manually editing any file in semantic/model/.

Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.

## MCP servers

- bigquery: transport: stdio; command: npx -y @toolbox-sdk/server --prebuilt bigquery --stdio; env: BIGQUERY_PROJECT
- cube: transport: streamable-http; url: ${CUBE_MCP_URL}

Read from the plugin's own mcp.json. Environment variable names only, never values.