databricks-aibi-dashboards
Pass
Audited by Gen Agent Trust Hub on Feb 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues were detected. The skill's instructions focus on technical implementation details for Databricks dashboard APIs.
- [COMMAND_EXECUTION]: The skill correctly uses specialized MCP tools like
execute_sqlandcreate_or_update_dashboardto perform its tasks. These tools are used as intended for validating queries and deploying workspace assets. - [PROMPT_INJECTION]: While the skill uses strong instructional language (e.g., "CRITICAL", "MANDATORY"), these are used to ensure the reliability of the dashboard generation process and do not attempt to bypass AI safety guardrails.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from Databricks tables (schemas and query results) to populate dashboard widgets. Although this is an external data source, the skill provides structured templates and strict validation rules that govern how this data is used, reducing the surface area for adversarial input to influence agent behavior.
Audit Metadata