M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
P

Pinecone MCP Server

Official
Connect your Pinecone projects to Cursor, Claude, and other AI assistants
How to install ↓ Compare GitHub source repository ↗
Category
Database #49 of 58
Stars
★ 71 Popular
Transport
stdio (local process)
Runtime
Node.js 18+
Credentials
Optional API key
License
Apache-2.0
Last commit
Tools
9
46FMRS · D

This MCP server is designed for developers to integrate Pinecone with AI coding assistants. It offers a rich toolset covering documentation search, index management, data operations, and advanced search capabilities. Installation is straightforward, supporting Cursor, Claude Desktop, and Gemini CLI. However, it is limited to indexes with integrated inference; those needing other features should consider alternatives.

Strongest · Setup experience 12/20 Weakest · Reliability 5/20

Reliability
5/20
Security and permissions
8/20
Maintenance
9/20
Documentation
12/20
Setup experience
12/20
Why each score
Reliability 5/20
This review only had the README and repository metadata; no source files, tests, or CI workflow files were available. Therefore we cannot verify that the server starts cleanly or completes the MCP init/tool-list handshake, nor that the declared tools match real behavior. The CI badge in the README is untrusted evidence. The 28 open issues add further uncertainty. Under the static calibration rule, reliability cannot exceed 12 without verifiable execution evidence; a conservative score of 5 is given.
Security and permissions 8/20
The only credential is PINECONE_API_KEY passed via environment variable, and the examples use placeholders rather than real tokens. No malware, credential theft, or covert exfiltration patterns are visible. The tools include side-effecting operations like create-index and upsert, but no destructive defaults (delete/pay/deploy/remote-command) without confirmation. However, there is no evidence of least-privilege scoping or explicit confirmation mechanisms server-side, and data-flow disclosure is incomplete. No red-line violation is present, so the score is 8.
Maintenance 9/20
The repository is not archived and uses the Apache-2.0 license, indicating basic ownership and a clear licensing path. However, the provided material contains no commit history, release cadence, dependency-update evidence, or security-response channel. 28 open issues suggest possible maintenance backlog. The CI claim in the README is untrusted, so a higher score is not justified; the score is 9.
Documentation 12/20
The README contains installation and configuration examples for Cursor, Claude desktop, and Gemini CLI, API key setup, a tool list, a clear limitation (only integrated-inference indexes are supported), and troubleshooting guidance. Missing are per-tool parameter details, cost information, and deeper usage examples. Some assumptions (e.g., network access to api.pinecone.io) are only implied in troubleshooting. Documentation is usable but has gaps, so the score is 12.
Setup experience 12/20
The README describes a clear, few-step setup: Node.js 18+ and npx are required, and JSON config examples are provided for Cursor, Claude desktop, and Gemini CLI. This is straightforward and covers multiple mainstream clients. However, the static calibration rule caps setup at 15 without verifiable execution evidence (real CI plus committed tests); we cannot confirm the npm package works or that the environment assumptions hold. A score of 12 reflects the clear instructions without execution proof.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessUses the network

Best for

  • Developers who use Pinecone as part of their technology stack.
  • Users who want to manage Pinecone projects and indexes from within AI coding assistants.
  • Those needing to interact with Pinecone documentation or generate code based on configuration.

Not for

  • Users working with indexes without integrated inference, or requiring standalone embeddings, vector search, or assistant features.
  • Scenarios where you need to provide AI assistants with context from your knowledge base (use Pinecone's Assistant MCP for that).
  • Non-developer use cases or non-coding assistant environments.

Required permissions

  • Requires a PINECONE_API_KEY environment variable to access your Pinecone project.
  • Allows AI tools to create, list, describe, and delete indexes (actions depend on tool set).
  • Allows upserting and searching records, as well as cross-index search and reranking operations.
  • Allows searching official Pinecone documentation (no API key needed).

Risks and side effects

  • API key exposure: ensure the key is kept secure and not committed to version control.
  • Misoperations: AI assistants may accidentally create, modify, or delete indexes; authorize carefully.
  • Cost implications: index operations and data usage may incur charges; monitor usage.
  • Network dependency: requires access to api.pinecone.io; may fail behind firewalls.

Setup

Before you start

Runtime：Node.js 18+

PINECONE_API_KEY optionalsecret Pinecone API key used to manage and query indexes; generate it in the Pinecone console at https://app.pinecone.io. Documentation search still works without it.

Generate a Pinecone API key (from console). Ensure Node.js v18 or later is installed and that node and npx are in your PATH.

Configure Cursor: Create a .cursor/mcp.json file in the project root with the configuration (see install_config). Check status in Cursor Settings > MCP.

Configure Claude Desktop: Locate claude_desktop_config.json via Settings > Developer > Edit Config, add the configuration, then restart.

Use as a Gemini CLI extension: Run gemini extensions install https://github.com/pinecone-io/pinecone-mcp and set the PINECONE_API_KEY environment variable.

.cursor/mcp.json 
{
  "mcpServers": {
    "pinecone": {
      "command": "npx",
      "args": [
        "-y",
        "@pinecone-database/mcp"
      ],
      "env": {
        "PINECONE_API_KEY": "<your pinecone api key>"
      }
    }
  }
}

Shown for Cursor. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json
{
  "servers": {
    "pinecone": {
      "command": "npx",
      "args": [
        "-y",
        "@pinecone-database/mcp"
      ],
      "env": {
        "PINECONE_API_KEY": "<your pinecone api key>"
      }
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal
claude mcp add pinecone -e 'PINECONE_API_KEY=<your pinecone api key>' -- npx -y @pinecone-database/mcp

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

Check the server status in Cursor Settings > MCP; once connected, tools like search-docs and list-indexes should appear, and the prompt 'List all my Pinecone indexes' should return results.

Troubleshooting

  1. MCP server not appearing in your tool: check Node.js version, npx in PATH, config file location and JSON syntax, and restart the tool.
  2. Invalid API key or auth errors: verify key in console, check environment variable, ensure no extra spaces or quotes.
  3. Tools not working: confirm index has integrated inference; otherwise create a new one with an embedding model. Check MCP server logs.
  4. Connection issues: ensure firewall allows api.pinecone.io, or run the server manually for detailed errors.

Things to try

Once connected, you can ask your AI assistant things like:

  • Search the Pinecone docs for information about metadata filtering
  • List all my Pinecone indexes and describe their configurations
  • Create a new index called 'my-docs' using the multilingual-e5-large model
  • Search my index for records related to 'authentication best practices'

Tools 9

search-docs read-only
Search the official Pinecone documentation.
list-indexes read-only
Lists all Pinecone indexes.
describe-index read-only
Describes the configuration of an index.
describe-index-stats read-only
Provides statistics about the data in the index, including the number of records and available namespaces.
create-index-for-model writes
Creates a new index that uses an integrated inference model to embed text as vectors.
upsert-records writes
Inserts or updates records in an index with integrated inference.
search-records read-only
Searches for records in an index based on a text query, using integrated inference for embedding. Has options for metadata filtering and reranking.
cascading-search read-only
Searches for records across multiple indexes, deduplicating and reranking the results.
Show 1 more tools
rerank-documents read-only
Reranks a collection of records or text documents using a specialized reranking model.

Use cases

Search Pinecone documentation within coding assistants for accurate answers.
Configure indexes based on your application's needs and generate code informed by your index configuration and data.
Test queries by upserting and searching data in your dev environment, evaluating results.
List all your indexes, describe their configurations, and inspect data statistics and namespaces.
Search records using integrated inference, apply metadata filtering and reranking.

Supported clients

Cursor
Claude Desktop
Gemini CLI

Listed from the project's documentation, not tested by this site.

Overview

The Pinecone Developer MCP Server enables AI assistants like Cursor, Claude Desktop, and Gemini CLI to interact with Pinecone projects and documentation. It supports searching documentation, managing indexes, performing data upserts and searches with integrated inference, and executing cross-index searches and reranking. The server is focused on improving the experience of developers working with Pinecone as part of their tech stack.

Similar servers

Compare with other MCP servers

Source revision a15d4b9ebc2e Data synced 2026-10-12 Read the FMRS scoring method