gomcptest
by owulveryck
gomcptest is a proof of concept demonstrating how to implement a Model Context Protocol (MCP) with a custom-built host for agentic systems. It provides tools for easy testing and iteration of specialized agents.
Last updated: N/A
What is gomcptest?
gomcptest is a proof-of-concept project showcasing how to build a custom host for the Model Context Protocol (MCP), enabling interaction with agentic systems. It includes a custom host mimicking the OpenAI API using Google Gemini and provides various MCP-compatible tools.
How to use gomcptest?
To use gomcptest, you need Go >= 1.21 and access to the Vertex AI API on Google Cloud Platform. Configure the environment variables in the .envrc file, build the tools using the Makefile, and run the openaiserver after navigating to its directory. You can then interact with the server using the OpenAI v1 chat completion format.
Key features of gomcptest
OpenAI Compatibility
Google Gemini Integration
Streaming Support
Function Calling
MCP Server Interaction
Single Chat Session
Use cases of gomcptest
Testing security vulnerability scanning agents
Creating code review agents for pull request analysis
Building data analysis agents for processing and visualizing complex datasets
Developing automated documentation agents
Rapid prototyping of agentic systems
Experimenting with custom tool integrations
FAQ from gomcptest
What is MCP?
What is MCP?
MCP stands for Model Context Protocol, a protocol for interacting with agentic systems.
What is the purpose of this project?
What is the purpose of this project?
The goal is to enable easy testing of agentic systems through the Model Context Protocol.
What are the prerequisites for running this project?
What are the prerequisites for running this project?
You need Go >= 1.21, access to the Vertex AI API on Google Cloud Platform, and github.com/mark3labs/mcp-go.
How do I configure the server?
How do I configure the server?
Configure the server using environment variables, as described in the Configuration section of the README.
Is this production-ready code?
Is this production-ready code?
No, this is a POC and has limitations. The code is provided as is for educational purposes.