skills.rest
pinecone-io
pinecone-ioOFFICIAL
Explore all skills by this author →
Website𝕏@pineconeGitHub

pinecone:quickstart

Guides users through creating a Pinecone index, upserting records, and running semantic search queries.

68|12|Updated Dec 17, 2025
Search Tags:#semantic search#embeddings#rag#vector database#mcp#quickstart#pinecone
One-click install
npx skills add https://github.com/pinecone-io/pinecone-claude-code-plugin --skill pinecone-quickstart-pinecone-io
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: pinecone:quickstart
Source: https://github.com/pinecone-io/pinecone-claude-code-plugin/tree/main/skills/quickstart
Command: npx skills add https://github.com/pinecone-io/pinecone-claude-code-plugin --skill pinecone-quickstart-pinecone-io

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pinecone, typer, and includes scripts (resource) components.

What problem does it solve? New Pinecone users face a steep onboarding curve: configuring API keys, choosing between vector database and Assistant workflows, creating indexes, and running their first semantic search. This Skill provides an interactive, guided walkthrough that takes users from zero to a working Pinecone setup. ## Core Features & Use Cases - Dual Learning Paths: Choose between the Database path (create an integrated index, upsert sample data, run semantic search with optional reranking) or the Assistant path (create a document Q&A assistant with cited answers). - Bundled Sample Data: Ships with 9 sample records across productivity, health, and nature themes so users can immediately see semantic search surface meaning rather than keyword matches. - Standalone Script Export: Copies a complete Python script (index creation, upsert, query, rerank) to the user's working directory for independent experimentation. - Use Case: A developer evaluating Pinecone runs the quickstart, creates a serverless index with integrated embeddings via the Pinecone MCP, upserts sample records, and runs natural-language queries to validate semantic search before committing to the platform. ## Quick Start Ask the assistant to walk you through the Pinecone quickstart so you can create your first index and run a semantic search.

Frequently Asked Questions about pinecone:quickstart

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I get started with Pinecone for the first time?▼

Run the Pinecone quickstart, which offers two guided paths: a Database path that creates an integrated index and runs semantic search, or an Assistant path that builds a document Q&A assistant. Set your PINECONE_API_KEY environment variable before starting.

How do I create a Pinecone index with integrated embeddings?▼

Use the Pinecone MCP create-index-for-model tool with a serverless configuration specifying cloud, region, and an embedding model like llama-text-embed-v2. The field_map parameter tells Pinecone which record field contains the text to embed.

What is the difference between Pinecone Database and Pinecone Assistant?▼

Pinecone Database is a vector store where you manage indexes, upsert records, and run queries yourself. Pinecone Assistant is a fully managed RAG service that handles chunking, embedding, and indexing automatically, returning cited answers to questions about uploaded documents.

Why is my Pinecone API key not working in Claude Code?▼

The PINECONE_API_KEY environment variable must be exported in your terminal before starting Claude Code, since it reads your shell environment. Alternatively, run scripts with uv run --env-file .env to load the key from a .env file, then restart your session.

Does Pinecone semantic search require matching keywords?▼

No, semantic search finds records by meaning rather than keyword overlap. The quickstart demonstrates this by querying with phrases that share no words with the stored records, yet still surfacing the correct themed results.

How do I improve Pinecone search result relevance with reranking?▼

Enable reranking in the search-records call by specifying a rerank model such as bge-reranker-v2-m3, the rank fields, and topN. Reranking runs a second-pass model over initial results to improve relevance ordering.

Related Skills

Explore More Software Engineering Skills→
pinecone-iopinecone-io
23

query

Search Pinecone indexes with natural language via MCP tools.

Official
Intermediate
twjarviszyra-webtwjarviszyra-web
3

pinecone

Build production vector search indexes for semantic retrieval workloads.

Community
Intermediate
Norfolk-GroupNorfolk-Group

pinecone-assistant

Create and manage Pinecone Assistants for cited Q&A over uploaded documents.

Official
Advanced
Norfolk-GroupNorfolk-Group

pinecone-quickstart

Create Pinecone integrated indexes, upsert sample records, and run semantic search queries.

Official
Intermediate