The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use execute_sql to validate queries. This is a standard and necessary function for the primary purpose of dashboard creation within the Databricks environment.
[DATA_EXPOSURE_SURFACE]: The skill interacts with database schemas through get_table_details and query results through execute_sql. This interaction is scoped to the Databricks platform and is essential for designing dashboard datasets.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests untrusted data from the database at runtime.
Ingestion points: Data enters the agent context through get_table_details (table schemas) and execute_sql (query results) as described in SKILL.md.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the database data.
Capability inventory: The skill uses execute_sql (data read/execution) and create_or_update_dashboard (write capability to the workspace).
Sanitization: There is no explicit logic described for sanitizing or filtering content retrieved from the database before it is used to structure the dashboard JSON.

databricks-aibi-dashboards