outlier-detection-analysis
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed for statistical analysis within the Observe platform. It uses a rigorous workflow for parameter resolution, ensuring that critical inputs like datasets and thresholds are confirmed by the user before query execution.
- [COMMAND_EXECUTION]: The skill generates complex OPAL queries that are executed within the host agent's controlled environment. The instructions include specific sanitization steps, such as casting values to strings and using sanitized column name suffixes, to prevent malformed query structures.
- [DATA_EXFILTRATION]: The skill operates on observability data (logs, spans, metrics) and does not attempt to access sensitive system files or credentials. Data processing is contained within the intended analytical use case.
- [PROMPT_INJECTION]: Instructional logic is focused on operational accuracy and does not contain patterns intended to bypass AI safety guidelines or override system prompts.
Audit Metadata