Back
AI summary
Written by AI from the official notes. Check them for exact details.The Google Antigravity SDK 0.1.1 enhances MCP integration, adds Vertex AI authentication, and improves error handling.
- Added support for allowlist and denylist in MCP configurations.
- Integrated native Vertex AI authentication in Python SDK.
- Automatic prefixing for MCP tools to prevent name collisions.
- Improved error handling with descriptive exceptions.
- Fixed token usage tracking for structured output.
Why it matters: Developers using the Antigravity SDK should update to benefit from improved functionality and error management.
Full release notes6 changes
This release focuses on significant enhancements to the Model Context Protocol (MCP) integration, adds native Vertex AI authentication support, improves robustness with better error handling, and includes critical fixes for token usage tracking.
- MCP Tool Filtering & Simplified Policies: Added support for
enabled_tools(allowlist) anddisabled_tools(denylist) in server configurations, and overloaded safety policy helpers (policy.allow,policy.deny,policy.ask_user) to accept the MCP server configuration object directly. - Vertex AI Authentication: Integrated native support for Vertex AI authentication in the Python SDK.
- MCP Tool Prefixing & Validation: The SDK now automatically namespaces and prefixes MCP tools (
mcp_{server_name}_{tool_name}) to prevent name collisions when connecting multiple MCP servers. Thenamefield is now mandatory in MCP server configurations and validated as a proper Python identifier. - Improved Error Handling: The SDK now raises explicit, descriptive exceptions for terminal errors rather than failing silently.
- Structured Output Token Tracking: Fixed a bug where token
usage_metadatawas not correctly returned when structured output (response.structured_output()) was requested. - Type Checking Warnings: Fixed
pytypewarnings (includingwrong-keyword-argsinLocalAgentConfig) across several modules.
Experience liftoff
Product
Resources