MCP Client (OpenAI and Claude)

This repo contains two separate MCP clients — one is MCP Client with OpenAI Integration and another one is with Claude Integration — that can connect to local MCP server and handle interactions. The client uses standard input/output (stdio) to talk to the MCP server.

---

Directory Structure

.
├── .git/
├── src/
│   ├── mcp-client-openai.ts
│   └── mcp-client-claude.ts
├── package.json
├── tsconfig.json
└── README.md

---

Setup Instructions

1. Clone the Repository

git clone https://github.com/devgotomarket/MCP-Client.git
cd MCP-Client

2. Install dependencies

npm install

3. Compile TypeScript

npm run build

This will generate the compiled JS files under the build/ directory.

---

4. prepare .env

Make sure your .env file is set up with the appropriate API key depending on the client:

# For OpenAI
OPENAI_API_KEY=your_openai_key_here

OPENAI_MODEL=openai_model ex:(gpt-4o)

# For Claude
ANTHROPIC_API_KEY=your_anthropic_key_here

ANTHROPIC_MODEL=anthropic_model ex:(claude-3-5-sonnet-20241022)

---

5. Run MCP Client (LLM-OpenAI)

node build/mcp-client-openai.js path/to/server.py

6. Run Claude Client (LLM-Claude)

node build/mcp-client-claude.js path/to/server.py

Replace path/to/server.py with the path to the MCP server script you want to connect to — it can be local or remote.

---

Notes

• You can use either .py (Python) or .js (Node) server scripts.
• The client will automatically detect the type and establish the transport.
• Tool results and LLM responses are streamed in an interactive prompt.

---