gemini-agents-api

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFE
Full Analysis
  • Official API Integration: The skill demonstrates how to interact with aiplatform.googleapis.com, which is the legitimate and authoritative global endpoint for Google Cloud AI services. These interactions are consistent with the skill's stated purpose of managing agent resources.
  • Standard Authentication Patterns: The instructions utilize Application Default Credentials (ADC) and the gcloud CLI to generate temporary access tokens. This is a recommended security practice for Google Cloud API interactions, as it avoids hardcoding sensitive credentials and relies on short-lived tokens.
  • Configurable Tool Capabilities: The skill allows for the provisioning of agents with specific toolsets, including code_execution and filesystem. While these provide significant capabilities to the managed agents, they are presented as standard platform features for user-defined configuration.
  • Secure Tool Integration (MCP): The documentation includes examples for Model Context Protocol (MCP) server configuration. It correctly uses placeholders for authentication tokens and includes specific guidance regarding the confidentiality of custom headers, ensuring sensitive information is only routed to the intended third-party endpoints.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 02:25 PM
Security Audit — agent-trust-hub — gemini-agents-api