mcp-registry/mcp-pinecone

    ==================
      
       /// MCP ///
      /// MCP ///
        
    ==================
        
    [server:online]
    [protocol:ready]

mcp-pinecone

by sirmews

Model Context Protocol (MCP) server that lets Claude Desktop (or any MCP client) read from and write to a Pinecone vector index. Provides basic RAG-style tools such as semantic-search, read-document, list-documents, pinecone-stats and process-document.

135
28
Open Source
GitHub

01

semantic-search

Search for records in the Pinecone index.

02

read-document

Read a document from the Pinecone index.

03

list-documents

List all documents in the Pinecone index.

04

pinecone-stats

Get stats about the Pinecone index, including the number of records, dimensions, and namespaces.

05

process-document

Process a document into chunks and upsert them into the Pinecone index, performing chunking, embedding, and upserting.

Installation

1. Clone the repository
git clone https://github.com/sirmews/mcp-pinecone.git && cd mcp-pinecone
2. Create an isolated Python environment (recommended)
python -m venv .venv && source .venv/bin/activate
3. Install Python dependencies
pip install -r requirements.txt
4. Export the required environment variables so the server can reach Pinecone
export PINECONE_API_KEY=<your-pinecone-key> export PINECONE_ENVIRONMENT=<us-east1-gcp / eu-west1-gcp / etc> # Optional – change default server port export MCP_PORT=8000
5. Initialise the Pinecone index (first-time only)
python scripts/init_index.py # <— if such helper script exists; otherwise create index in dashboard>
6. Start the MCP server
python -m mcp_pinecone # or uvicorn mcp_pinecone.server:app --host 0.0.0.0 --port ${MCP_PORT:-8000}
7. Verify it is running
curl http://localhost:8000/health

Documentation

License: MIT License
Updated 7/15/2025