skills/googlecloudplatform/data-agent-kit-plugin/gcp-managed-airflow-recommendations/Gen Agent Trust Hub
gcp-managed-airflow-recommendations
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
gcloudCLI tool to obtain authentication tokens and retrieve environment configuration data. - Evidence: The
_run_gcloudhelper inscripts/lib/monitoring.pyexecutesgcloudviasubprocess.run. Additionally,references/gcloud_reference.mdprovides commands for describing environments and reading logs. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze external data, such as Cloud Composer logs and environment metrics, which could potentially contain malicious instructions.
- Ingestion points: Data is ingested from Cloud Composer logs (using
gcloud logging read) and the Cloud Monitoring PromQL API as described inSKILL.mdand thescripts/directory. - Boundary markers: There are no explicit delimiters or specific instructions for the agent to ignore potentially malicious embedded content within the logs.
- Capability inventory: The skill can execute local commands (
gcloud) and perform network operations to Google Cloud API endpoints (monitoring.googleapis.com). - Sanitization: The skill does not implement sanitization or filtering for the external log content before it is processed by the agent.
Audit Metadata