What makes Valyu AI different from standard search engines?

It is specifically optimized for AI agents, providing LLM-ready markdown, structured JSON extraction, and detailed metadata like citations and relevance scores.

Does this skill support academic research?

Yes, the API can search proprietary datasets and provides DOIs, author names, and citation counts for scholarly results.

Is streaming supported for AI answers?

Yes, the Answer API supports Server-Sent Events (SSE) to stream search results and AI-generated content in real-time for a faster user experience.

How do I extract specific data from a website?

You can use the Contents API with a JSON schema to define the exact structure you want the AI to extract from any given URL.

Can I control the cost of search queries?

Absolutely. Valyu AI includes parameters like max_price and relevance_threshold to help manage CPM-based spending directly within the skill.

Valyu AI Search & Research

Name: Valyu AI Search & Research
Author: lev-os

bylev-os

0•

网络抓取与数据收集

Provides real-time web search, AI-grounded answers, and automated content extraction optimized for AI agents.

Valyu AI is an enterprise-grade search and research skill designed specifically for AI agents, offering real-time access to web data, proprietary datasets, and news. It goes beyond simple search by providing grounded AI answers, structured content extraction via JSON schemas, and deep research capabilities with full citation support. This skill is ideal for developers building autonomous agents that require accurate, up-to-date information with fine-grained control over sources, costs, and output formats, making it a powerful tool for RAG and automated discovery workflows.

主要功能

01Grounded AI Answer API with support for streaming and custom system instructions

02Advanced source filtering to include or exclude specific domains, news sources, or academic papers

03Structured content extraction from URLs using custom JSON schemas and markdown output

04Deep research capabilities for automated, multi-turn information gathering and reporting

050 GitHub stars

06Real-time web and proprietary dataset search with customizable relevance thresholds

使用场景

01Extracting structured product or pricing data from web pages for competitive analysis

02Powering RAG systems with real-time news and web context for up-to-date AI responses

03Building autonomous research agents that require academic citations and DOI tracking

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add lev-os/agents search-valyu

For use in Claude.ai and ChatGPT

Download Skill