skills.rest
github
githubOFFICIAL
Explore all skills by this author →
WebsiteGitHub

pinecone-rag

Build RAG pipelines and persistent agent memory using Pinecone vector database.

38.5k|4.9k|Updated Jun 11, 2025
Search Tags:#embeddings#vector database#agent memory#pinecone#rAG#Semantic Search#Hybrid Search
One-click install
npx skills add https://github.com/github/awesome-copilot --skill pinecone-rag
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: pinecone-rag
Source: https://github.com/github/awesome-copilot/tree/main/skills/pinecone-rag
Command: npx skills add https://github.com/github/awesome-copilot --skill pinecone-rag

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pinecone, openai, pinecone-text.

What problem does it solve?

Connecting an LLM to a searchable knowledge base requires correct index configuration, embedding pipelines, and retrieval strategies — mistakes like dimension mismatches or missing namespaces cause silent failures and cross-tenant data leaks. This Skill provides a step-by-step workflow for building production RAG systems and persistent agent memory on Pinecone.

Core Features & Use Cases

  • Index Setup & Embedding Pipeline: Create serverless or pod-based indexes with correct dimensions, then batch-embed and upsert documents with text stored in metadata.
  • Flexible Retrieval Strategies: Choose dense semantic search, hybrid search combining BM25 with dense vectors, or metadata-filtered queries for scoped results.
  • Agent Memory with Namespace Isolation: Store, recall, and delete per-agent memories using namespaces to prevent cross-user data leaks.
  • Use Case: A team wants their support chatbot to answer questions from internal documentation. Use this Skill to index the docs into Pinecone, retrieve relevant chunks per query, and ground the LLM's answers in that context.

Quick Start

Ask the AI to build a RAG pipeline that indexes your documents into Pinecone and answers questions using retrieved context.

Frequently Asked Questions about pinecone-rag

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

FAQPage Schema
How do I build a RAG pipeline with Pinecone?▼

Create a Pinecone index with dimensions matching your embedding model, batch-embed and upsert documents with text stored in metadata, then query for relevant chunks and inject them into the LLM's system prompt as context.

How do I store agent memory across sessions with a vector database?▼

Embed each memory and upsert it into a per-agent namespace like agent_<id>, then recall memories by querying that namespace semantically. Namespaces isolate each agent's memories and allow full deletion with delete_all on request.

When should I use hybrid search instead of dense semantic search in Pinecone?▼

Use hybrid search when your corpus contains exact terminology that semantic search misses, such as legal citations, medical codes, product SKUs, or API method names. Combine BM25 sparse vectors with dense vectors using an alpha weight.

Why does my Pinecone upsert fail with a dimension mismatch?▼

The index dimension must exactly match your embedding model's output, such as 1536 for text-embedding-3-small or 1024 for voyage-3. Verify len(embed(["test"])[0]) equals the index dimension before your first upsert.

When should I not use Pinecone for vector search?▼

Avoid Pinecone when the dataset fits in memory (use FAISS or Chroma), when you already run PostgreSQL (use pgvector), or when you need sub-5ms p99 latency without external API calls. It is a managed cloud service, not a local store.

Related Skills

Explore More Software Engineering Skills→
HosungYouHosungYou
5

i3

Download PDFs, extract text, chunk documents, and create a local ChromaDB vector database.

Community
Advanced
choice5346choice5346

langchain

Build LLM-powered applications with agents, chains, and RAG capabilities.

Community
Advanced
Giskard-AIGiskard-AI
5

rag-evaluator

Evaluate RAG systems for groundedness, relevance, and citation accuracy.

Official
Advanced
JhonHanderJhonHander

langchain-rag

Construct RAG pipelines with OpenAI embeddings and FAISS, Chroma, or Pinecone vector stores.

Community
Advanced