P

pinecone.io

Pinecone provides a managed vector database and related AI retrieval services, including inference and document-grounded assistant features. Developers use it to store embeddings, search and rank results, and build retrieval-augmented applications.

7 integrations · MCP · REST · CLI

Pinecone exposes one public REST API, three MCP servers (marketing, docs, and per-assistant remote MCP), and a `pc` CLI; auth is via project API keys, service-account OAuth client credentials for admin access, or interactive CLI login.

discovered 1d ago
MCP servers5
REST · OpenAPI1
CLI1
Credentials
Pinecone project API keyapi_keyGet key ↗

Create a project API key in the Pinecone console or follow the steps in Manage API keys: open your project, go to API keys, click Create API key, choose permissions, and copy the key value. Send it as the Api-Key header for HTTP APIs, or configure it with pc auth configure --api-key / PINECONE_API_KEY for the CLI.

Pinecone service account client credentialsoauth2_ccSet up OAuth ↗

Create a service account in the Pinecone console as described in Manage service accounts at the organization-level, then add it to a project via project-level service accounts. Copy the client_id and client_secret, then exchange them for an access token at https://login.pinecone.io/oauth/token using the client_credentials grant with audience https://api.pinecone.io/. The resulting bearer token authenticates Admin API requests; the raw client credentials can also be configured directly in the CLI with pc auth configure --client-id ... --client-secret ... or PINECONE_CLIENT_ID / PINECONE_CLIENT_SECRET.

Pinecone user login token for CLIoauth2$ pc auth login

Acquired by the CLI — running pc auth login opens the auth flow and stores the credential.

conventions · 4/8 published

Publish these signals → /publishing