Google Vertex AI Search

What is mcp-vertexai-search?

mcp-vertexai-search is a MCP (Multi-Cloud Platform) server designed to facilitate document searching using Google Cloud's Vertex AI Search. It leverages Gemini models with Vertex AI grounding to enhance search results by grounding responses in your private data stored in Vertex AI Datastore. This allows for more accurate and relevant search outcomes by integrating with your specific datasets.

How to use mcp-vertexai-search?

There are two primary ways to use this MCP server:

1. Clone the repository:

# Clone the repository
git clone git@github.com/ubie-oss/mcp-vertexai-search.git

# Create a virtual environment
uv venv
# Install the dependencies
uv sync --all-extras

# Check the command
uv run mcp-vertexai-search

2. Install the Python package:

While not yet on PyPI, you can install it directly from the repository. A config.yml file (derived from config.yml.template) is required to run the server.

# Install the package
pip install git+https://github.com/ubie-oss/mcp-vertexai-search.git

# Check the command
mcp-vertexai-search --help

To run the MCP server, use the following command, specifying the configuration file and transport method (stdio or sse):

uv run mcp-vertexai-search serve \
    --config config.yml \
    --transport <stdio|sse>

You can also test the Vertex AI Search directly without the MCP server:

uv run mcp-vertexai-search search \
    --config config.yml \
    --query <your-query>

Key Features of mcp-vertexai-search

• Vertex AI Grounding: Improves search result quality by grounding Gemini's responses in your private data stored in Vertex AI Datastore.
• Integration with Multiple Data Stores: Capable of integrating one or multiple Vertex AI data stores.
• Flexible Deployment: Can be run via Docker (Dockerfile provided) or as a Python package.
• Configurable: Allows for detailed configuration via a YAML file (config.yml) for server settings, model parameters, and data store details.
• Supports SSE and Stdio Transports: Offers flexibility in how the server communicates.

Use Cases of mcp-vertexai-search

• Internal Knowledge Bases: Companies can build powerful internal search tools that leverage their proprietary documents and data.
• Customer Support Systems: Enhance chatbots and support portals with accurate, grounded information from product documentation or FAQs.
• Research and Development: Facilitate efficient searching of research papers, patents, or internal R&D documents.
• Data-driven Applications: Develop applications that require precise information retrieval from large, private datasets.

FAQ from mcp-vertexai-search

Q: What is Vertex AI Grounding? A: Vertex AI Grounding is a feature that enhances the quality of search results by using your private data stored in Vertex AI Datastore to ground the responses generated by a Gemini model. This ensures that the search results are more relevant and accurate to your specific context.

Q: How do I configure the MCP server? A: The MCP server is configured using a YAML file, typically named config.yml. A template, config.yml.template, is provided, which outlines parameters for the server name, Vertex AI model details (name, project ID, location, service account), and a list of Vertex AI data stores (project ID, location, datastore ID, tool name, description).

Q: Can I use multiple Vertex AI data stores? A: Yes, the solution is designed to integrate with one or multiple Vertex AI data stores, allowing you to search across various datasets.

Q: Is the Python package available on PyPI? A: Currently, the package is not published to PyPI. You can install it directly from the GitHub repository.

MCP Server for Vertex AI Search

This is a MCP server to search documents using Vertex AI.

Architecture

This solution uses Gemini with Vertex AI grounding to search documents using your private data. Grounding improves the quality of search results by grounding Gemini's responses in your data stored in Vertex AI Datastore. We can integrate one or multiple Vertex AI data stores to the MCP server. For more details on grounding, refer to Vertex AI Grounding Documentation.

How to use

There are two ways to use this MCP server. If you want to run this on Docker, the first approach would be good as Dockerfile is provided in the project.

1. Clone the repository

# Clone the repository
git clone git@github.com:ubie-oss/mcp-vertexai-search.git

# Create a virtual environment
uv venv
# Install the dependencies
uv sync --all-extras

# Check the command
uv run mcp-vertexai-search

Install the python package

The package isn't published to PyPI yet, but we can install it from the repository. We need a config file derives from config.yml.template to run the MCP server, because the python package doesn't include the config template. Please refer to Appendix A: Config file for the details of the config file.

# Install the package
pip install git+https://github.com/ubie-oss/mcp-vertexai-search.git

# Check the command
mcp-vertexai-search --help

Development

Prerequisites

• uv
• Vertex AI data store
  • Please look into the official documentation about data stores for more information

Set up Local Environment

# Optional: Install uv
python -m pip install -r requirements.setup.txt

# Create a virtual environment
uv venv
uv sync --all-extras

Run the MCP server

This supports two transports for SSE (Server-Sent Events) and stdio (Standard Input Output). We can control the transport by setting the --transport flag.

We can configure the MCP server with a YAML file. config.yml.template is a template for the config file. Please modify the config file to fit your needs.

uv run mcp-vertexai-search serve \
    --config config.yml \
    --transport <stdio|sse>

Test the Vertex AI Search

We can test the Vertex AI Search by using the mcp-vertexai-search search command without the MCP server.

uv run mcp-vertexai-search search \
    --config config.yml \
    --query <your-query>

Appendix A: Config file

config.yml.template is a template for the config file.

• server
  • server.name: The name of the MCP server
• model
  • model.model_name: The name of the Vertex AI model
  • model.project_id: The project ID of the Vertex AI model
  • model.location: The location of the model (e.g. us-central1)
  • model.impersonate_service_account: The service account to impersonate
  • model.generate_content_config: The configuration for the generate content API
• data_stores: The list of Vertex AI data stores
  • data_stores.project_id: The project ID of the Vertex AI data store
  • data_stores.location: The location of the Vertex AI data store (e.g. us)
  • data_stores.datastore_id: The ID of the Vertex AI data store
  • data_stores.tool_name: The name of the tool
  • data_stores.description: The description of the Vertex AI data store