configuring-audit-logging
Pass
Audited by Gen Agent Trust Hub on Jul 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
ccloudCLI tool to retrieve cluster information and verify log export configurations on CockroachDB Cloud (ccloud cluster info <cluster-name> -o json). This is a legitimate management operation within the context of the vendor's ecosystem. - [PROMPT_INJECTION]: The skill defines an attack surface for indirect prompt injection via the use of placeholders such as
<role_name>and<cluster-name>in SQL and CLI commands. - Ingestion points: User-provided inputs for cluster names and role names in
SKILL.mdandreferences/sql-queries.md. - Boundary markers: Absent.
- Capability inventory: SQL execution for cluster setting modifications (
SET CLUSTER SETTING) and shell execution via theccloudtool. - Sanitization: No specific validation or escaping mechanisms are described for the interpolated variables.
Audit Metadata