configuring-log-export
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines procedures that ingest user-provided configuration strings—such as cluster IDs, IAM ARNs, and API keys—and interpolates them directly into shell commands and SQL queries. This pattern represents a surface for potential indirect prompt injection if the user inputs are maliciously crafted.
- Ingestion points: User-provided inputs for
<cluster-id>,<cluster-name>,<iam-role-arn>,<aws-region>,<gcp-project-id>,<cockroachdb-service-account>, and<datadog-api-key>inSKILL.md. - Boundary markers: None present; the instructions do not use specific delimiters or protective framing for the interpolated variables.
- Capability inventory: The skill instructions utilize shell execution capabilities via the
ccloud,aws, andgcloudCLI tools, as well as SQL execution capabilities for cluster settings (SKILL.md). - Sanitization: The instructions do not specify any validation or sanitization steps for the user-supplied variables before they are executed in the shell or database context.
Audit Metadata