What is Vertex Ai Mcp Server

The vertex-ai-mcp-server is a Model Context Protocol (MCP) server that facilitates interaction with Google Cloud’s Vertex AI Gemini models, primarily focusing on coding assistance and general query answering.

Use cases

Use cases include answering coding-related queries, finding official documentation for specific topics, and providing general knowledge answers using the capabilities of the Vertex AI models.

How to use

To use the vertex-ai-mcp-server, clone the project repository, install the necessary dependencies using ‘bun install’, and configure your environment variables for model settings and Google Cloud authentication.

Key features

Key features include access to Vertex AI Gemini models through various MCP tools, support for web search grounding and direct knowledge answering, configurable model parameters, streaming API for improved responsiveness, and basic retry logic for transient errors.

Where to use

The vertex-ai-mcp-server can be used in fields such as software development for coding assistance, customer support for query answering, and any domain requiring intelligent interaction with AI models.

Vertex AI MCP Server

This project implements a Model Context Protocol (MCP) server that provides a comprehensive suite of tools for interacting with Google Cloud’s Vertex AI Gemini models, focusing on coding assistance and general query answering.

Features

• Provides access to Vertex AI Gemini models via numerous MCP tools.
• Supports web search grounding (answer_query_websearch) and direct knowledge answering (answer_query_direct).
• Configurable model ID, temperature, streaming behavior, max output tokens, and retry settings via environment variables.
• Uses streaming API by default for potentially better responsiveness.
• Includes basic retry logic for transient API errors.
• Minimal safety filters applied (BLOCK_NONE) to reduce potential blocking (use with caution).

Tools Provided

Query & Generation (AI Focused)

• answer_query_websearch: Answers a natural language query using the configured Vertex AI model enhanced with Google Search results.
• answer_query_direct: Answers a natural language query using only the internal knowledge of the configured Vertex AI model.
• explain_topic_with_docs: Provides a detailed explanation for a query about a specific software topic by synthesizing information primarily from official documentation found via web search.
• get_doc_snippets: Provides precise, authoritative code snippets or concise answers for technical queries by searching official documentation.
• generate_project_guidelines: Generates a structured project guidelines document (Markdown) based on a specified list of technologies (optionally with versions), using web search for best practices.

Research & Analysis Tools

• code_analysis_with_docs: Analyzes code snippets by comparing them with best practices from official documentation, identifying potential bugs, performance issues, and security vulnerabilities.
• technical_comparison: Compares multiple technologies, frameworks, or libraries based on specific criteria, providing detailed comparison tables with pros/cons and use cases.
• architecture_pattern_recommendation: Suggests architecture patterns for specific use cases based on industry best practices, with implementation examples and considerations.
• dependency_vulnerability_scan: Analyzes project dependencies for known security vulnerabilities, providing detailed information and mitigation strategies.
• database_schema_analyzer: Reviews database schemas for normalization, indexing, and performance issues, suggesting improvements based on database-specific best practices.
• security_best_practices_advisor: Provides security recommendations for specific technologies or scenarios, with code examples for implementing secure practices.
• testing_strategy_generator: Creates comprehensive testing strategies for applications or features, suggesting appropriate testing types with coverage goals.
• regulatory_compliance_advisor: Provides guidance on regulatory requirements for specific industries (GDPR, HIPAA, etc.), with implementation approaches for compliance.
• microservice_design_assistant: Helps design microservice architectures for specific domains, with service boundary recommendations and communication patterns.
• documentation_generator: Creates comprehensive documentation for code, APIs, or systems, following industry best practices for technical documentation.

Filesystem Operations

• read_file_content: Read the complete contents of one or more files. Provide a single path string or an array of path strings.
• write_file_content: Create new files or completely overwrite existing files. The ‘writes’ argument accepts a single object ({path, content}) or an array of such objects.
• edit_file_content: Makes line-based edits to a text file, returning a diff preview or applying changes.
• list_directory_contents: Lists files and directories directly within a specified path (non-recursive).
• get_directory_tree: Gets a recursive tree view of files and directories as JSON.
• move_file_or_directory: Moves or renames files and directories.
• search_filesystem: Recursively searches for files/directories matching a name pattern, with optional exclusions.
• get_filesystem_info: Retrieves detailed metadata (size, dates, type, permissions) about a file or directory.
• execute_terminal_command: Execute a shell command, optionally specifying cwd and timeout. Returns stdout/stderr.

Combined AI + Filesystem Operations

• save_generate_project_guidelines: Generates project guidelines based on a tech stack and saves the result to a specified file path.
• save_doc_snippet: Finds code snippets from documentation and saves the result to a specified file path.
• save_topic_explanation: Generates a detailed explanation of a topic based on documentation and saves the result to a specified file path.
• save_answer_query_direct: Answers a query using only internal knowledge and saves the answer to a specified file path.
• save_answer_query_websearch: Answers a query using web search results and saves the answer to a specified file path.

(Note: Input/output schemas for each tool are defined in their respective files within src/tools/ and exposed via the MCP server.)

Prerequisites

• Node.js (v18+)
• Bun (npm install -g bun)
• Google Cloud Project with Billing enabled.
• Vertex AI API enabled in the GCP project.
• Google Cloud Authentication configured in your environment (Application Default Credentials via gcloud auth application-default login is recommended, or a Service Account Key).

Setup & Installation

1. Clone/Place Project: Ensure the project files are in your desired location.
2. Install Dependencies:

   bun install
   

3. Configure Environment:
   • Create a .env file in the project root (copy .env.example).
   • Set the required and optional environment variables as described in .env.example.
     • Set AI_PROVIDER to either "vertex" or "gemini".
     • If AI_PROVIDER="vertex", GOOGLE_CLOUD_PROJECT is required.
     • If AI_PROVIDER="gemini", GEMINI_API_KEY is required.
4. Build the Server:

   bun run build
   

   This compiles the TypeScript code to build/index.js.

Usage (Standalone / NPX)

Once published to npm, you can run this server directly using npx:

# Ensure required environment variables are set (e.g., GOOGLE_CLOUD_PROJECT)
bunx vertex-ai-mcp-server

Alternatively, install it globally:

bun install -g vertex-ai-mcp-server
# Then run:
vertex-ai-mcp-server

Note: Running standalone requires setting necessary environment variables (like GOOGLE_CLOUD_PROJECT, GOOGLE_CLOUD_LOCATION, authentication credentials if not using ADC) in your shell environment before executing the command.

Installing via Smithery

To install Vertex AI Server for Claude Desktop automatically via Smithery:

bunx -y @smithery/cli install @shariqriazz/vertex-ai-mcp-server --client claude

Running with Cline

1. Configure MCP Settings: Add/update the configuration in your Cline MCP settings file (e.g., .roo/mcp.json). You have two primary ways to configure the command:

   Option A: Using Node (Direct Path - Recommended for Development)

   This method uses node to run the compiled script directly. It’s useful during development when you have the code cloned locally.

   • Important: Ensure the args path points correctly to the build/index.js file. Using an absolute path might be more reliable.

   Option B: Using NPX (Requires Package Published to npm)

   This method uses npx to automatically download and run the server package from the npm registry. This is convenient if you don’t want to clone the repository.

   • Ensure the environment variables in the env block are correctly set, either matching .env or explicitly defined here. Remove comments from the actual JSON file.

2. Restart/Reload Cline: Cline should detect the configuration change and start the server.

3. Use Tools: You can now use the extensive list of tools via Cline.

Development

• Watch Mode: bun run watch
• Linting: bun run lint
• Formatting: bun run format

License

This project is licensed under the MIT License - see the LICENSE file for details.