bigquery-ai-ml
Pass
Audited by Gen Agent Trust Hub on Jun 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill, associated with author kilo-org, provides instructional content for BigQuery machine learning functions (AI.FORECAST, AI.DETECT_ANOMALIES, and AI.GENERATE) hosted in a trusted repository.
- [SAFE]: All external dataset references point to official Google Cloud public datasets and sample storage buckets.
- [PROMPT_INJECTION]: Documentation in references/ai_generate.md contains examples of indirect prompt injection surfaces where database records are interpolated into model prompts. (1) Ingestion points: references/ai_generate.md (dataset.articles); (2) Boundary markers: Absent; (3) Capability inventory: AI.GENERATE function calls; (4) Sanitization: Not present in educational SQL examples. This follows standard BigQuery ML documentation patterns for the primary purpose of the skill.
Audit Metadata