Build on the supported package contract
Use @absolutejs/rag-pinecone through its supported public entry points.
@absolutejs/rag-pineconev0.0.13betaAIPinecone vector-store adapter for @absolutejs/rag
bun add @absolutejs/rag-pineconePinecone vector-store adapter for @absolutejs/rag. Treats Pinecone as the backend boundary: server-side filtering, native vector search, and explicit opt-in index provisioning helpers.
Index provisioning
The Pinecone index is not auto-provisioned at runtime. For explicit, opt-in provisioning use the exported helpers:
Outcomes
Use @absolutejs/rag-pinecone through its supported public entry points.
Hardening checklist
Follow in order
Working example for Usage.
import { createPineconeRAG } from '@absolutejs/rag-pinecone';
const rag = createPineconeRAG({
apiKey: process.env.PINECONE_API_KEY,
indexName: 'absolute-rag-demo',
namespace: 'production',
vector: {
provider: 'pinecone',
dimensions: 1536,
distanceMetric: 'cosine'
}
});
await rag.store.upsert({
chunks: [
{
chunkId: 'doc-1#0',
text: 'Pinecone stores vectors with attached metadata.',
title: 'Pinecone overview',
source: 'https://docs.pinecone.io',
metadata: { tags: ['vector', 'managed'] }
}
]
});
const hits = await rag.collection.search({ query: 'vector database', topK: 4 });The Pinecone index is not auto-provisioned at runtime. For explicit, opt-in provisioning use the exported helpers:
import {
describePineconeIndex,
ensurePineconeIndex
} from '@absolutejs/rag-pinecone';
await ensurePineconeIndex({
apiKey: process.env.PINECONE_API_KEY,
indexName: 'absolute-rag-demo',
dimensions: 1536,
metric: 'cosine',
waitUntilReady: true
});Supported entry points declared by this package manifest.
Package entry point declared in package.json.
Scripts declared by this package manifest.