Dev Local Vector Store for Genkit
This is a simple implementation of a vector store that can be used to local development and testing.
Building with a coding agent? Install the Genkit JS skill first.
npx skills add genkit-ai/skills --skill developing-genkit-jsIt teaches your agent the current Genkit JS APIs and common gotchas. Source, manual install and skills for other languages: genkit-ai/skills.
This plugin is not meant to be used in production.
Installing the plugin
npm i --save @genkit-ai/dev-local-vectorstore
Using the plugin
import { Document, genkit } from 'genkit';
import { googleAI } from '@genkit-ai/google-genai';
import {
devLocalVectorstore,
devLocalIndexerRef,
devLocalRetrieverRef,
} from '@genkit-ai/dev-local-vectorstore';
const ai = genkit({
plugins: [
googleAI(),
devLocalVectorstore([
{
indexName: 'BobFacts',
embedder: googleAI.embedder('gemini-embedding-001'),
},
]),
],
model: googleAI.model('gemini-flash-latest'),
});
// Reference to a local vector database storing Genkit documentation
const indexer = devLocalIndexerRef('BobFacts');
const retriever = devLocalRetrieverRef('BobFacts');
async function main() {
// Add documents to the index. Only do it once.
await ai.index({
indexer: indexer,
documents: [
Document.fromText('Bob lives on the moon.'),
Document.fromText('Bob is 42 years old.'),
Document.fromText('Bob likes bananas.'),
Document.fromText('Bob has 11 cats.'),
],
});
const question = 'How old is Bob?';
// Consistent API to retrieve most relevant documents based on semantic similarity to query
const docs = await ai.retrieve({
retriever: retriever,
query: question,
});
const result = await ai.generate({
prompt: `Use the provided context from the Genkit documentation to answer this query: ${question}`,
docs, // Pass retrieved documents to the model
});
console.log(result.text);
}
main();
The sources for this package are in the main Genkit repo. Please file issues and pull requests against that repo.
Usage information and reference details can be found in official Genkit documentation.
License: Apache 2.0