pinecone-mcp-server

dario-suckfuell/pinecone-mcp-server

3.1

If you are the rightful owner of pinecone-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.

Pinecone MCP Server is a minimal remote MCP server with SSE transport designed for OpenAI Deep Research.

Pinecone MCP Server

Minimal remote MCP server with SSE transport for OpenAI Deep Research.

What You Need

  • Python 3.10+
  • Pinecone index (1536 dimensions)
  • OpenAI API key

Setup

# Install (requires Python 3.10+)
pip install -r requirements.txt

# Test tools work
python test_tools.py

# Run server
python pinecone_mcp.py

Configuration

Edit env file with your keys:

PINECONE_API_KEY=your_key
PINECONE_INDEX=your_index
PINECONE_HOST=your_host
OPENAI_API_KEY=your_key

Deploy to Railway

  1. Push to GitHub
  2. Connect on railway.app
  3. Add environment variables from env file
  4. Deploy

Use with Deep Research

from openai import OpenAI

client = OpenAI()
response = client.chat.completions.create(
    model="o4-mini-deep-research",
    messages=[{"role": "user", "content": "Search my database"}],
    tools=[{"type": "mcp", "mcp": {"url": "https://your-server.railway.app"}}]
)

Files

  • pinecone_mcp.py - Main server (search + fetch tools)
  • test_tools.py - Test your setup works
  • env - Configuration
  • requirements.txt - Dependencies
  • Procfile - Deploy config

That's it. See GUIDE.md for more details if needed.