mcp-vertex-search-snippets v0.1.2 Released: Critical Bug Fix for US Location Users
Introduction
On September 3, 2025, we released version v0.1.2 of mcp-vertex-search-snippets, a targeted bug fix release that resolves a critical connection issue affecting users with US-based Vertex AI Search configurations.
This release addresses an API endpoint URL construction problem that prevented users from successfully connecting to the Vertex AI Search service when configured with location: "us". If you upgraded to v0.1.1 and experienced unexpected connection failures, this release restores full functionality.
What’s New
Critical Bug Fix: US Location URL Construction
The headline fix in v0.1.2 corrects how the application constructs the Google Cloud Discovery Engine API endpoint for US location users.
The Problem: In v0.1.1, the code inadvertently created a malformed domain URL (usus-discoveryengine.googleapis.com) that caused connection failures for anyone using location: "us" in their configuration.
The Solution: v0.1.2 now correctly generates the proper domain endpoint (us-discoveryengine.googleapis.com), restoring full connectivity for US-based deployments.
What This Means For You
- US Location Users: If you have
location: "us"in yourvertex.yamlconfiguration and experienced connection failures after upgrading to v0.1.1, this fix resolves the issue immediately. - EU and Global Users: No impact – configurations using
location: "eu"orlocation: "global"were unaffected.
Why It Matters
This release ensures that mcp-vertex-search-snippets works reliably across all supported Google Cloud regions. The Model Context Protocol server is designed to integrate seamlessly with Vertex AI Search for document discovery and extractive snippet retrieval. Connection failures due to URL construction errors would have completely blocked this functionality for affected users.
The fix maintains full backward compatibility – no changes to your configuration files or setup are required. Simply upgrade to v0.1.2, and the correct endpoint will be used automatically.
Upgrade Information
Installing v0.1.2
From Source:
go install github.com/UnitVectorY-Labs/mcp-vertex-search-snippets@v0.1.2
Pre-compiled Binaries:
Download the appropriate binary for your platform from the v0.1.2 release page:
- macOS (Intel):
mcp-vertex-search-snippets-v0.1.2-darwin-amd64.tar.gz - macOS (Apple Silicon):
mcp-vertex-search-snippets-v0.1.2-darwin-arm64.tar.gz - Linux: Available for AMD64, 386, and ARM64 architectures
- Windows: Available for AMD64 and 386
All release assets include checksums (MD5 and SHA256) for verification.
Configuration
No configuration changes are needed. Your existing vertex.yaml file works as-is:
project_id: "your-project-id"
location: "us" # Now working correctly
app_id: "your-engine-id"
Looking Ahead
While v0.1.2 resolves the US location domain issue, users may want to consider upgrading to v0.1.3 (released October 8, 2025), which includes additional improvements to the API path version (/v1alpha/ → /v1/) for enhanced compatibility with Google Cloud’s stable API endpoints.
Transparency Note: This release announcement was AI-generated using the unsloth/Qwen3.5-122B-A10B-GGUF:Q4_K_M model on March 17, 2026. Source material: UnitVectorY-Labs/mcp-vertex-search-snippets, Release v0.1.2 (September 3, 2025). Authored by release-storyteller.