Happy slice of mcp pizzaMCP.pizza

gomcptest

MCP.Pizza Chef: owulveryck

gomcptest is a proof-of-concept MCP client that implements a custom host with an OpenAI-compatible API, integrating Google Vertex AI, function calling, and interaction with MCP servers. It is designed to facilitate easy testing and development of agentic systems, such as code review agents, security scanners, and data analysis tools, by providing a clear, from-scratch implementation of MCP mechanisms.

Use This MCP client To

Test agentic systems with custom MCP client implementations Integrate Google Vertex AI with OpenAI-compatible APIs Develop code review agents analyzing pull requests Create security vulnerability scanning agents for codebases Build data analysis agents for complex datasets Automate documentation generation from source code

README

gomcptest: Proof of Concept for MCP with Custom Host

This project is a proof of concept (POC) demonstrating how to implement a Model Context Protocol (MCP) with a custom-built host to play with agentic systems. The code is primarily written from scratch to provide a clear understanding of the underlying mechanisms.

See the experimental website for documentation (auto-generated) at https://owulveryck.github.io/gomcptest/

Goal

The primary goal of this project is to enable easy testing of agentic systems through the Model Context Protocol. For example:

  • The dispatch_agent could be specialized to scan codebases for security vulnerabilities
  • Create code review agents that can analyze pull requests for potential issues
  • Build data analysis agents that process and visualize complex datasets
  • Develop automated documentation agents that can generate comprehensive docs from code

These specialized agents can be easily tested and iterated upon using the tools provided in this repository.

Prerequisites

  • Go >= 1.21
  • Access to the Vertex AI API on Google Cloud Platform
  • github.com/mark3labs/mcp-go

The tools use the default GCP login credentials configured by gcloud auth login.

Project Structure

diagram

  • host/openaiserver: Implements a custom host that mimics the OpenAI API, using Google Gemini and function calling. This is the core of the POC.
  • tools: Contains various MCP-compatible tools that can be used with the host:
    • Bash: Execute bash commands
    • Edit: Edit file contents
    • GlobTool: Find files matching glob patterns
    • GrepTool: Search file contents with regular expressions
    • LS: List directory contents
    • Replace: Replace entire file contents
    • View: View file contents

gomcptest FAQ

How does gomcptest integrate with Google Vertex AI?
gomcptest acts as a custom MCP client that connects to Google Vertex AI using an OpenAI-compatible API, enabling function calling and model interaction.
Can gomcptest be used to build specialized agents?
Yes, it supports creating agents for code review, security scanning, data analysis, and automated documentation.
Is gomcptest suitable for production use?
gomcptest is a proof-of-concept primarily for experimentation and learning, not optimized for production environments.
Where can I find documentation for gomcptest?
Documentation is auto-generated and available at https://owulveryck.github.io/gomcptest/.
Does gomcptest support function calling?
Yes, it includes function calling capabilities to extend agent interactions.
How does gomcptest interact with MCP servers?
It communicates with MCP servers to fetch context and execute multi-step reasoning workflows.
What programming languages is gomcptest written in?
The project is primarily written from scratch, focusing on clarity of MCP implementation, typically in Go language.
Can gomcptest be extended to other LLM providers?
While focused on Google Vertex AI, its OpenAI-compatible API design allows adaptation to other providers like OpenAI, Anthropic, and Claude.