Skip to main content

Quick Reference

Minimal steps to enable the ragSearch tool for grounded retrieval.

Set your key for embeddings:

export OPENAI_KEY=sk-...

2) Configure​

import { ReactAgentBuilder } from "delreact-agent";

const agent = new ReactAgentBuilder({
  geminiKey: process.env.GEMINI_KEY,
  openaiKey: process.env.OPENAI_KEY,
  rag: {
    vectorFiles: [
      "./example/asset/attention-is-all-you-need.json",
      "./example/asset/color-palletes-in-marketing.json"
    ],
    embeddingModel: "text-embedding-3-small",
    topK: 5,
    threshold: 0.7
  }
})
.init({ selectedProvider: "gemini", model: "gemini-2.5-flash" })
.build();

ragSearch is auto-registered and available when openaiKey and rag.vectorFiles are provided.

3) Use​

Agent-driven (recommended):

const result = await agent.invoke({
  objective: "Summarize core ideas from 'Attention Is All You Need' with citations"
});
console.log(result.conclusion);

Force a direct tool call (advanced):

import { toolRegistry } from "delreact-agent/core/tools";

const tools = toolRegistry.createToolsWithConfig(agent.config);
const rag = tools.find(t => t.name === "ragSearch");
const output = rag ? await rag.invoke({ query: "Role of self-attention?" }) : null;

Tips​

  • Install hnswlib-node for faster search on larger corpora.
  • Tune threshold (e.g., 0.6–0.8) to filter low-relevance hits.
  • Provide multiple vectorFiles to aggregate sources.