The Agent Skills Directory

[SAFE]: The skill is a legitimate resource for defining and querying business metrics in Databricks. It aligns with official platform documentation and uses standard patterns for Metric Views.
[EXTERNAL_DOWNLOADS]: The skill references official documentation and resources from docs.databricks.com. These are well-known and trusted service domains used for providing technical guidance on Databricks features.
[COMMAND_EXECUTION]: The skill utilizes the manage_metric_views tool to perform administrative actions such as create, alter, drop, and grant. These commands are standard for managing data assets within the Databricks environment and are used for their intended purpose.
[INDIRECT_PROMPT_INJECTION]: The skill involves processing SQL expressions provided within YAML structures for dimensions and measures.
Ingestion points: Data enters the system via the dimensions, measures, and filter fields in the YAML definition and tool arguments.
Boundary markers: Definitions are encapsulated within SQL string delimiters ($$) or structured Python tool arguments.
Capability inventory: The skill has the ability to manage view life-cycles and user permissions through the manage_metric_views tool.
Sanitization: While the skill does not explicitly mention sanitization of SQL expressions, it is designed for users with appropriate workspace permissions to define governed business logic.

databricks-metric-views