What is the 'toon' output format?

The 'toon' format is a YAML-like output designed for high readability and token efficiency, offering a 40% reduction in tokens compared to standard JSON.

How much does it cost to retrieve content?

Exa.ai typically charges $0.001 per piece of content (per URL). The skill encourages token optimization strategies to keep your overall AI costs low.

Can I extract specific fields like prices or dates from a website?

Yes, by using the --summary-schema parameter, you can define exactly which data points you want the AI to extract into a structured JSON object.

How does Exa Get Contents save on LLM tokens?

The skill utilizes AI-generated summaries and structured schema extraction to filter out web noise, reducing token usage by up to 90% compared to raw HTML scraping.

Does this skill work with dynamic or JavaScript-heavy websites?

Yes, it uses a mandatory live-crawl feature with a configurable timeout to ensure that even dynamic content is captured and processed correctly.

Exa Get Contents

Name: Exa Get Contents
Author: benjaminjackson

bybenjaminjackson

•

ウェブスクレイピングとデータ収集

Extracts structured data and AI-generated summaries from any URL with high token efficiency and live crawling.

概要

Exa Get Contents is a specialized Claude Code skill designed to retrieve and process web content using Exa's AI-native search engine. It enables developers to perform precise web scraping, summarize complex articles, and extract specific information into structured JSON formats. By leveraging features like live crawling for fresh data and 'toon' output for token reduction, it provides a cost-effective way to feed high-quality, relevant web data into Claude's context for analysis, research, or content generation tasks.

主な機能

2 GitHub stars
AI-powered content summarization with custom natural language queries
Granular control over extraction limits for text, links, and image metadata
Mandatory live crawling to ensure access to real-time, non-cached web content
Structured data extraction using custom JSON Schema definitions
Token-optimized output formats including 'toon' and JSON for efficient processing

ユースケース

Generating concise technical summaries of long-form documentation or news articles to save context space
Extracting specific product features and pricing from competitor landing pages into structured JSON
Automating data collection for RAG pipelines by retrieving clean, noise-free text from multiple URLs

概要

主な機能

2 GitHub stars
AI-powered content summarization with custom natural language queries
Granular control over extraction limits for text, links, and image metadata
Mandatory live crawling to ensure access to real-time, non-cached web content
Structured data extraction using custom JSON Schema definitions
Token-optimized output formats including 'toon' and JSON for efficient processing

ユースケース

Generating concise technical summaries of long-form documentation or news articles to save context space
Extracting specific product features and pricing from competitor landing pages into structured JSON
Automating data collection for RAG pipelines by retrieving clean, noise-free text from multiple URLs