observability
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch grammar and command documentation from the official opensearch-project/sql GitHub repository. This is a vendor-owned resource used for query validation.
- [COMMAND_EXECUTION]: Provides examples for using curl to interact with OpenSearch APIs and mentions a local script for searching documentation. The curl examples use the -k flag which bypasses SSL certificate verification.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze log and trace data from OpenSearch clusters, which may contain untrusted content.
- Ingestion points: Data retrieved from OpenSearch indices via PPL queries.
- Boundary markers: None specified in the instructions for delimiting log content.
- Capability inventory: PPL query execution and local script interaction.
- Sanitization: Not explicitly mentioned in the provided instructions.
Audit Metadata