skills/timsonner/agent-skills/gcloud/Gen Agent Trust Hub

gcloud

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides bash commands for interacting with Google Cloud Platform services using the official gcloud CLI. These commands are used for legitimate discovery, reconnaissance, and vulnerability analysis of GCP projects, IAM policies, and storage resources.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines project and identity parameters to scope its operations. 1. Ingestion points: The 'Project, account' argument-hint in the SKILL.md frontmatter. 2. Boundary markers: The procedure includes a mandatory step to 'Confirm the active account and project context' before executing queries. 3. Capability inventory: The skill leverages gcloud capabilities for accessing project metadata, IAM policies, storage buckets, and secret manager values. 4. Sanitization: The instructions include safeguards to 'Query only the services relevant to the current hypothesis' and 'Default to read-only validation', reducing the risk of unintended or excessive data access.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 02:39 AM
Security Audit — agent-trust-hub — gcloud