Summary

It reads its own docs before it writes your code.

search-docs is the point of this being Pinecone's own server: the assistant grounds itself in the current documentation before generating index config or query code, rather than relying on what a model happens to remember. The upsert and search tools then let it test those queries against your real data in the dev loop — but note they only work on integrated-inference indexes, so describe-index is the check to run first.

What it is

Pinecone's official developer server, aimed at coding assistants: 9 tools for searching Pinecone's documentation, discovering and creating indexes, and upserting and searching records against your project. Runs over stdio.

What you get

search-docs searches Pinecone's official documentation — the tool the assistant is meant to hit before writing Pinecone code rather than guessing at behavior
Index discovery: list-indexes shows every index with its config and readiness, describe-index returns the embedding model and the fieldMap that names the text field to embed, and describe-index-stats reports record counts per namespace so you can see which namespaces exist
create-index-for-model stands up an integrated-inference index that embeds the named text field automatically, across AWS, GCP or Azure, and blocks until the index is ready so there is no readiness polling afterwards
upsert-records inserts or updates records in an integrated index, embedding the fieldMap text automatically and overwriting any record with an existing id
search-records finds semantically similar records in an integrated index (it embeds the query for you) and can rerank in the same call, while cascading-search runs across several indexes then deduplicates and reranks the combined results
rerank-documents reorders externally sourced text by relevance to a query, for documents that did not come from a Pinecone search

Requirements

Node.js v20 or later with node and npx on PATH, run as npx @pinecone-database/mcp. A Pinecone API key in PINECONE_API_KEY from app.pinecone.io — without it the assistant can still search the documentation, but cannot manage or query your indexes. The search and upsert tools work on integrated-inference indexes, which embed text automatically; describe-index tells you an index's type.

Setup effort

One command plus a key — npx @pinecone-database/mcp, then supply credentials