Prerequisites

1. Have Python 3.10 or higher installed
2. Preferably having uv installed
   1. Steps to install uv https://docs.astral.sh/uv/getting-started/installation/
3. Use node 22+ NPM 10.0.0 or higher
   1. Steps to install NPM https://docs.npmjs.com/downloading-and-installing-node-js-and-npm
4. Install a MCP Client
   1. Claude Desktop: https://claude.ai/download
   2. Continue on VSCode:Link for Continue on VSCode
   3. Cursor: Link for Cursor MCP Instructions

Local MCP Server

Install dependencies

Create virtual environment with uv for python 3.11 and install dependencies using uv pip

uv python install 3.11
uv venv --python 3.11
uv .venv/bin/pip install -r requirements.txt

Env File

AZURE_OPENAI_API_KEY= The API key for your Azure OpenAI service
AZURE_OPENAI_ENDPOINT= The endpoint for your Azure OpenAI service
AZURE_OPENAI_EMBEDDING_DEPLOYMENT= The deployment name for your Azure OpenAI embedding service
AZURE_OPENAI_DEPLOYMENT= The deployment name for your Azure OpenAI model
AZURE_OPENAI_EMBEDDING_VERSION= The version of the Azure OpenAI embedding service

Build the local context

[Optional] Run the notebook to build the local context.The notebook is located at notebook/langchain_tool.ipynb

[Optional] The file notebook/sklearn_vectorstore.parquet is what is generated by the notebook and is used by the server for vector data.

Replace Value in the Server file

Replace the PATH value in paychex-mcp.py with the path to your local context

PATH = "<PATH_TO_YOUR_REPO>/notebook/"

MCP Inspector

Run the mcp inspector locally

The MCP inspector is a developer tool for testing and debugging MCP servers.

bash-3.2$ npx @modelcontextprotocol/inspector

Usage

Add the env values and arguments to your MCP Inspector command

bash-3.2$  npx @modelcontextprotocol/inspector \
  <path_to_repo>/.venv/bin/python \
  /<path_to_repo>/paychex-mcp.py \
  -e AZURE_OPENAI_API_KEY=<API_KEY> \
  -e AZURE_OPENAI_ENDPOINT=<ENDPOINT> \
  -e AZURE_OPENAI_EMBEDDING_DEPLOYMENT=<EMBEDDING_DEPLOYMENT> \
  -e AZURE_OPENAI_DEPLOYMENT=<DEPLOYMENT> \
  -e AZURE_OPENAI_EMBEDDING_VERSION=<EMBEDDING_VERSION>

MCP Inspector Flags

You can pass both arguments and environment variables to your MCP server. Arguments are passed directly to your server, while environment variables can be set using the -e flag:

Pass arguments only

npx @modelcontextprotocol/inspector build/index.js arg1 arg2

Pass environment variables only

npx @modelcontextprotocol/inspector -e KEY=value -e KEY2=$VALUE2 node build/index.js

Pass both environment variables and arguments

npx @modelcontextprotocol/inspector -e KEY=value -e KEY2=$VALUE2 node build/index.js arg1 arg2

Use -- to separate inspector flags from server arguments

npx @modelcontextprotocol/inspector -e KEY=$VALUE -- node build/index.js -e server-flag

Use MCP Inspector with other MCP Servers

npx -y @modelcontextprotocol/inspector npx @modelcontextprotocol/server-filesystem arg1 agr2

Using MCP Configuration files

The mcp-config folder contains configuration files necessary to set up and run MCP (Model Context Protocol) servers for various environments. Here are the steps to utilize these configuration files effectively:

Step 1: Locate the MCP Configuration Files

Navigate to the mcp-config folder in your project directory. Inside, you will find JSON configuration files for different setups, such as:

• claude_desktop_config.json for Claude Desktop setup.
• config.yaml for continue VS Code Extension setup.

Step 2: Edit Configuration Files

Before using the configuration files, you might need to edit them to match your local or deployment environments. For example, in claude_desktop_config.json or config.yaml, replace placeholders such as <path to your Python executable>, <your Azure OpenAI API key>, and others with actual values relevant to your setup.

MCP Server list

MCP Server List Filesystem, Git, Atlassian, etc. servers Filesystem server Git server Atlassian server