Valyu — Unified Data Search for AI Agents

A single API for 25+ specialized data sources: academic papers, SEC filings, clinical trials, patent databases, real-time news, and prediction markets. Built for AI agents that need reliable, citable research data.

Integrates natively with Claude, OpenAI, LangChain, LlamaIndex, and the Vercel AI SDK. $10 free credits to get started at platform.valyu.ai.

What This Skill Does

Core Capabilities

Search API: Query across web, academic (40M+ papers via arXiv/PubMed), medical, financial, and proprietary sources with date range and relevance filters
Contents API: Clean markdown extraction from up to 10 URLs simultaneously; supports structured JSON schema extraction
Answer API: AI-synthesized responses with citations; fast mode for low latency, streaming support
DeepResearch API: Comprehensive research reports in fast (~5 min), standard (~10–20 min), and heavy (~90 min) tiers
27 Recipes: Ready-to-use workflows covering search, extraction, answers, and research

Installation

BASH

npx skills add valyuAI/skills

Get your API key at platform.valyu.ai.

Quick Start

PYTHON

from valyu import Valyu

client = Valyu(api_key="your_api_key")

# Search academic papers
results = client.search(
    query="CRISPR gene editing cancer treatment efficacy",
    sources=["academic", "medical"],
    max_results=10
)

# Get a synthesized answer with citations
answer = client.answer(
    question="What are the latest developments in quantum computing?",
    mode="fast"
)