Datasources API | TypeScript SDK

A tool manifest for AI agents to discover available data sources at runtime, then pass their ids into Search, Answer, or DeepResearch via includedSources.

Discover Valyu data sources at runtime with the TypeScript SDK.

Open in Cursor

Basic usage

import { Valyu } from "valyu-js";

const valyu = new Valyu();

// List datasources (optionally filtered by category)
const response = await valyu.datasources.list({ category: "research" });
const sourceIds = (response.datasources ?? []).map((ds) => ds.id);

// Feed them straight into a search
await valyu.search("latest transformer architecture improvements", { includedSources: sourceIds });

// List categories with dataset counts
const { categories } = await valyu.datasources.categories();
categories?.forEach((cat) => console.log(cat.id, cat.name, cat.dataset_count));

example_queries on each datasource doubles as few-shot examples for prompting, and pricing.cpm lets you estimate cost before searching.

Category	Description
`research`	Academic papers (arXiv, PubMed, bioRxiv)
`healthcare`	Clinical trials, drug info, health data
`markets`	Stocks, crypto, forex, ETFs
`company`	SEC filings, earnings, insider trades
`economic`	FRED, BLS, World Bank data
`predictions`	Polymarket, Kalshi
`transportation`	UK Rail, ship tracking
`legal`	Case law, legislation
`politics`	Parliamentary data
`patents`	Global patent filings

Reference

Methods

Method	Description
`datasources.list({ category? })`	List datasources, optionally filtered by category
`datasources.categories()`	List all categories with dataset counts

Response format

interface DatasourcesListResponse {
  success: boolean;
  error?: string;
  datasources?: Datasource[];
}

interface Datasource {
  id: string;                              // e.g., "valyu/valyu-arxiv"
  name: string;
  description: string;
  category: DatasourceCategoryId;
  type: string;
  modality: DatasourceModality[];
  topics: string[];
  languages?: string[];
  source?: string;
  example_queries: string[];               // Sample queries for few-shot prompting
  pricing: DatasourcePricing;              // .cpm = cost per million tokens
  response_schema?: Record<string, any>;
  update_frequency?: string;
  size?: number;
  coverage?: DatasourceCoverage;           // .start_date / .end_date
}

type DatasourceCategoryId =
  | "research" | "healthcare" | "patents" | "markets" | "company"
  | "economic" | "predictions" | "transportation" | "legal" | "politics";

type DatasourceModality = "text" | "images" | "tabular";

interface DatasourcesCategoriesResponse {
  success: boolean;
  error?: string;
  categories?: DatasourceCategory[];
}

interface DatasourceCategory {
  id: string;
  name: string;
  description: string;
  dataset_count: number;
}

​Basic usage

​Categories

​Reference

Basic usage

Categories

Reference