Create and configure Databricks Asset Bundles for multi-environment deployments.
npx skills add https://github.com/FMurray/mlfts --skill asset-bundles-fmurray
Please help me install this Agent Skill. Skill: asset-bundles Source: https://github.com/FMurray/mlfts/tree/main/.agents/skills/asset-bundles Command: npx skills add https://github.com/FMurray/mlfts --skill asset-bundles-fmurray
databricks.yml with variables, targets, and includes following the standard project structure. - Resource Configuration: Provides correct YAML patterns for dashboards, pipelines, jobs, SQL alerts (v2 API), volumes, and apps, including permission levels and grants. - Multi-Environment Deployment: Parameterizes catalog, schema, and warehouse via variables so the same bundle deploys to dev and prod with different profiles. - Use Case: You need to add a dashboard and a scheduled job to a new bundle and deploy to prod. The Skill supplies the resource YAML with correct ../src/ paths, permission levels, and the databricks bundle deploy -t prod command. ## Quick Start Ask the agent to create a Databricks Asset Bundle with a dashboard and a pipeline, parameterized for dev and prod targets.High-intent search queries and answers about installing and using this skill.
Define a databricks.yml with variables for catalog, schema, and warehouse, then create targets for dev and prod with different workspace profiles and variable overrides. Deploy with databricks bundle deploy -t prod to target a specific environment.
Add a dashboards resource in a resources/*.yml file with display_name, file_path pointing to the .lvdash.json file, warehouse_id, and dataset_catalog/dataset_schema parameters. Use ../src/ paths since resource files sit one level below the bundle root.
Paths are relative to the file that declares them: resources/*.yml files must use ../src/ to reach the bundle root, while databricks.yml uses ./src/. Mixing these formats is the most common cause of path resolution failures.
Dashboards support CAN_READ, CAN_RUN, CAN_EDIT, and CAN_MANAGE; jobs support CAN_VIEW, CAN_MANAGE_RUN, and CAN_MANAGE. Volumes use grants instead of permissions, and the admins group cannot be modified on jobs.
The Alert v2 API uses evaluation with comparison_operator, source, and threshold instead of a condition field, and schedule requires pause_status, quartz_cron_schedule, and timezone_id directly. Run databricks bundle schema to inspect the exact AlertV2 structure before writing YAML.
Define an apps resource with name, description, and source_code_path, and put environment variables in app.yaml inside the source directory rather than databricks.yml. After deploying, start the app with databricks bundle run <resource_key> and debug with databricks apps logs.
JingyiWngCreate and configure Databricks Declarative Automation Bundles for multi-environment deployments.
alsk1992Spawn, monitor, and control child processes with configurable resource limits.
Zack2626-okAutomate creation and management of Databricks Spark Declarative Pipelines.
partme-aiGuide Terraform infrastructure as code with HCL syntax and state management.