google-antigravity-sdk

Pass

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a documentation repository for the Google Antigravity SDK. It provides instructional content and code examples for using the SDK's features.
  • [SAFE]: Includes explicit security guidance regarding 'Safety Policies' (references/safety_policies.md), documenting how to restrict or permit agent capabilities like file system access and shell command execution using a priority-based policy model.
  • [SAFE]: Recommends secure credential management practices, such as using environment variables (GEMINI_API_KEY) or .env files rather than hardcoding secrets.
  • [SAFE]: Describes integrations with the Model Context Protocol (MCP) and custom Python tools, providing clear examples for both local (Stdio) and remote (SSE) transport configurations.
  • [SAFE]: Provides mechanisms for monitoring and observability, including token usage tracking and custom tracing via lifecycle hooks.
Audit Metadata
Risk Level
SAFE
Analyzed
May 19, 2026, 08:13 PM
Security Audit — agent-trust-hub — google-antigravity-sdk