cargo-observability
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on the cargo-ai CLI tool to perform all observability tasks, including creating, listing, and previewing alerts. The instructions provide numerous examples of shell commands executed via this CLI.\n- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the @cargo-ai/cli package from npm. This package is a verified vendor resource belonging to the author, getcargohq.\n- [INDIRECT_PROMPT_INJECTION]: The skill provides an interface for executing custom SQL queries against orchestration runtime tables and data warehouses through the orchestrationQuery and storageQuery scope types.\n
- Ingestion points: SQL queries are provided via the --scope argument in the cargo-ai observability alert create and preview commands.\n
- Boundary markers: None identified for user-provided SQL content.\n
- Capability inventory: SQL execution via cargo-ai against workspace databases (ClickHouse).\n
- Sanitization: No explicit sanitization or validation of SQL queries is mentioned in the instruction files; however, the service requires specific observability:read and observability:write permissions.
Audit Metadata