Skip to content
AI Dev Toolkit.
Esc
  • AI Token CounterCount tokens for GPT, Claude, Gemini, DeepSeek, Qwen and more.Tool
  • LLM API Cost CalculatorEstimate per-request, daily and monthly API costs.Tool
  • AI Model ComparisonCompare prices, context windows and features across models.Tool
  • AI Model Pricing PagesSpecs, real costs and cheaper alternatives for popular models.Tool
  • Context Window CheckerSee whether your text fits each model’s context window.Tool
  • Subscription vs API CalculatorFind out whether a chat plan or the API is cheaper for you.Tool
  • GPU / VRAM CalculatorCheck how much VRAM a local model needs and which GPUs fit.Tool
  • Prompt Caching CalculatorEstimate savings from prompt caching.Tool

MCP server · Databases

Pinecone Developer MCP server

Lets a coding assistant search Pinecone docs, create and inspect indexes, and upsert, search and rerank records while you build.

Maintainer
Pinecone (official)
Licence
Apache-2.0
Runs
Locally (npx)
Sign-in
Optional token
Last activity
2026-08-14

Overview

What it does

Pinecone’s developer MCP server runs locally through npx and uses a Pinecone API key from the PINECONE_API_KEY environment variable. Without a key it can still search the Pinecone documentation, but it cannot touch your indexes. It is aimed at coding assistants, so the agent can see your index configuration and data while it writes code.

With a key, the assistant can list and describe indexes, read record counts per namespace, create an index that uses an integrated embedding model, upsert records and run text searches with metadata filters and reranking. A cascading search tool queries several indexes and merges the results. It only works with indexes that use integrated inference; standalone embeddings and plain vector search are not supported. Pinecone offers a separate Assistant MCP for giving assistants context from a knowledge base.

Tools

What the assistant can do

  • search-docs over the Pinecone documentation
  • list-indexes, describe-index and describe-index-stats
  • create-index-for-model with integrated inference
  • upsert-records and search-records with filtering and reranking
  • cascading-search across several indexes
  • rerank-documents

Install

Install Pinecone Developer MCP server in your app

You need:

  • Node.js 20 or newer
  • A Pinecone API key for index tools (docs search works without one)

Run in your project. --scope local keeps the servers private to you in this project; use user for every project.

Terminal
claude mcp add --env PINECONE_API_KEY="${PINECONE_API_KEY}" --scope local pinecone -- npx -y @pinecone-database/mcp
Native Windows (PowerShell)
claude mcp add --env PINECONE_API_KEY="$env:PINECONE_API_KEY" --scope local pinecone -- cmd /c npx -y '@pinecone-database/mcp'
  • Set PINECONE_API_KEY first, e.g. export PINECONE_API_KEY=… in your shell. The command saves the value, not the variable.

Want it alongside other servers? Open Pinecone Developer MCP Server in the MCP config generator and add the rest; it writes one file per app. The snippets follow the setup in the server’s README, checked on 2026-10-14.

Before you install

  • With an API key the assistant can create indexes and write records in that project; there is no read-only flag.
  • Use a key for a development project rather than production.
  • Only indexes with integrated inference are supported.
  • A local MCP server runs with your user account’s permissions. Install it only from the source linked above. See how to choose a safe server.

Why it’s listed

Why we list it

Official and focused on the developer loop of designing, filling and testing a Pinecone index.

Related

All MCP serversMCP config generatorGuide: MCP servers in Claude CodeGuide: what is MCP?