managing-google-maps-platform
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
gcloudCLI to perform administrative tasks on Google Cloud Platform. Specifically, it executes commands to list enabled services, audit API key restrictions, and query monitoring time-series data for usage metrics. These commands are consistent with the skill's stated purpose of managing Google Maps Platform resources. - [INDIRECT_PROMPT_INJECTION]: The skill processes output from external CLI tools (
gcloudandjq), creating an indirect prompt injection surface where the agent ingests untrusted data from the GCP environment. However, the risk is mitigated by the skill's read-only discovery-first pattern and explicit anti-hallucination rules that require verification of resource names before action. - [DATA_EXPOSURE_EXFILTRATION]: While the skill accesses sensitive metadata regarding API keys (such as existence and restrictions), it does so to perform a security audit (identifying unrestricted keys). There are no network operations targeting external or untrusted domains to exfiltrate this information.
Audit Metadata