gcp-cloud-functions
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes standard
gcloudCLI commands (such asgcloud functions list,gcloud functions describe, andgcloud monitoring time-series list) along with thejqutility to interact with Google Cloud Platform services. These operations are aligned with the skill's stated purpose of managing and diagnosing Cloud Functions.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests potentially untrusted data by reading application logs from Google Cloud Functions.\n - Ingestion points: Function logs are retrieved using the
gcloud functions logs readcommand inSKILL.md.\n - Boundary markers: The instructions do not define specific delimiters to separate log content from agent instructions, nor do they include warnings to ignore instructions embedded within the logs.\n
- Capability inventory: The skill has capabilities to discover, describe, and inspect metrics for Cloud Functions and Cloud Builds.\n
- Sanitization: Content is processed via
jqto filter for specific fields, but the textual content of the logs is presented to the agent for analysis.
Audit Metadata