OfficialVerified

Pinecone MCP Server

Query vector indexes and search Pinecone documentation from your AI assistant.

Data last scanned today · Reviewed today

639 tools★ 71TypeScriptApache-2.0

Overview

The Pinecone MCP Server bridges your AI coding tools with your Pinecone projects. By configuring this server, you enable your assistant to directly interact with your vector infrastructure, allowing it to provide more context-aware support for your application data and indexing configurations. Beyond managing indexes, the tool provides a direct conduit to Pinecone documentation. This helps ensure that the generated code and configuration advice are grounded in official examples and current guidance.

Our verdict

This server carries a high-risk profile with a security score of 40/100. Because critical components like authentication methods and read-only mode behavior remain unreviewed, users should exercise caution when granting it access to production API keys. It is best suited for development and testing environments rather than production workflows.

Supports active development with commits as recent as four days ago.
Maintains a compact supply chain with only three direct dependencies.
Includes documentation for querying and upserting data into indexes.
Licensed under the permissive Apache-2.0 open source license.

Security review for authentication and read-only modes is currently missing.
Lacks an official vendor maintenance status despite its utility.
Community-maintained status increases long-term support uncertainty.

Setup

Claude Desktop

{
  "mcpServers": {
    "pinecone-mcp": {
      "args": [
        "-y",
        "@pinecone-database/mcp"
      ],
      "command": "npx"
    }
  }
}

Paste into ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows), then restart Claude Desktop. Some servers require API keys or tokens as environment variables — check the project's README and add an "env" object if needed.

No read-only mode — running this grants full read/write access.
No built-in authentication — anyone with access to this config can invoke every tool it exposes.

Tools

Tool	Description	Risk
search-docs	Search the official Pinecone documentation.	low
list-indexes	Lists all Pinecone indexes.	low
describe-index	Describes the configuration of an index.	low
describe-index-stats	Provides statistics about the data in the index, including the number of records and available namespaces.	low
create-index-for-model	Creates a new index that uses an integrated inference model to embed text as vectors.	medium
upsert-records	Inserts or updates records in an index with integrated inference.	medium
search-records	Searches for records in an index based on a text query, using integrated inference for embedding.	low
cascading-search	Searches for records across multiple indexes, deduplicating and reranking the results.	low
rerank-documents	Reranks a collection of records or text documents using a specialized reranking model.	low

Compatibility

Client	Local	Docker	Remote	Read-only
Claude Desktop
Cursor
VS Code
Windsurf
ChatGPT

Frequently asked questions

›What are the software prerequisites for running this MCP server?

You must have Node.js version 20 or later installed, and both 'node' and 'npx' must be available in your system PATH.

›Do I need a Pinecone API key to use this server?

An API key is required to manage or query your indexes. However, you can still search official Pinecone documentation without providing one.

›What types of Pinecone indexes are supported?

This server exclusively supports indexes configured with integrated inference. Standard indexes, standalone embeddings, and legacy vector search configurations are not supported.

›How can I configure the server for Cursor?

Add a configuration block for the server within the 'mcpServers' object inside your project-level or global '.cursor/mcp.json' file.

›How do I install the server for the Gemini CLI?

You can install it as an extension by running 'gemini extensions install https://github.com/pinecone-io/pinecone-mcp' and setting your API key via the PINECONE_API_KEY environment variable.