databricks-aibi-dashboards
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill employs official Databricks CLI subcommands such as 'lakeview create' and 'experimental aitools' for legitimate dashboard management and data exploration.
- [SAFE]: No hardcoded secrets, sensitive path exposures, or unauthorized network activity were detected.
- [SAFE]: The instructions do not contain prompt injections, persistence mechanisms, or dynamic execution of untrusted remote code.
- [INDIRECT_PROMPT_INJECTION]: The skill interacts with Databricks workspace data to inform dashboard design. 1) Ingestion points: Results from 'databricks experimental aitools tools query' and 'discover-schema' (SKILL.md). 2) Boundary markers: None defined for data processing. 3) Capability inventory: 'lakeview' and 'workspace' subcommands (SKILL.md). 4) Sanitization: None mentioned. This functional surface is necessary for the skill's primary purpose and is assessed as safe.
Audit Metadata