What is the cost of running an agent built with this skill?

The architecture is designed to be completely free, utilizing GitHub Actions for compute, the Gemini Flash free tier for AI processing, and Notion's free tier for data storage.

How does the AI enrichment work?

The agent sends collected data to Gemini Flash in optimized batches to summarize, categorize, and score each item based on your specific criteria and priorities.

Can it scrape websites that require JavaScript rendering?

Yes, the skill provides patterns for using Playwright, which allows the agent to handle dynamic, JS-heavy websites that simple HTML scrapers cannot access.

Where is the scraped data stored?

The skill supports multiple storage backends including Notion, Google Sheets, and Supabase, as well as local JSON or CSV files.

How does the agent learn from my feedback?

The agent tracks your interactions with the data in a feedback loop, storing positive and negative signals in a JSON file to bias the AI's scoring logic for future runs.

AI Data Scraper Agent

Name: AI Data Scraper Agent
Author: affaan-m

byaffaan-m

•

172,009

•

Web Scraping & Data Collection

Builds automated, AI-powered data collection agents that scrape, enrich, and store information from any public source using free-tier tools.

This skill enables Claude to architect and implement production-ready data scraping pipelines using a robust three-layer architecture: Collect, Enrich, and Store. It provides implementation patterns for scraping HTML, APIs, and RSS feeds, then uses free-tier AI models like Gemini Flash to summarize and score the data. Designed for zero-cost operation, it utilizes GitHub Actions for scheduling and integrates seamlessly with Notion, Google Sheets, or Supabase for data management. It also features a unique feedback-loop system that allows the agent to learn from user decisions to improve its filtering accuracy over time.

Key Features

01Zero-cost deployment and scheduling via GitHub Actions

02Automated AI enrichment using Gemini Flash with built-in rate limiting

03Adaptive feedback system that learns user preferences for improved scoring

04Seamless storage integration with Notion, Google Sheets, and Supabase

05Multi-source scraping templates for HTML, REST APIs, and RSS feeds

06172,009 GitHub stars

Use Cases

01Tracking product prices and sending alerts for significant price drops

02Aggregating and classifying technical news or GitHub releases by sentiment and topic

03Monitoring job boards to find and score roles matching a specific resume

Key Features

01Zero-cost deployment and scheduling via GitHub Actions

02Automated AI enrichment using Gemini Flash with built-in rate limiting

03Adaptive feedback system that learns user preferences for improved scoring

04Seamless storage integration with Notion, Google Sheets, and Supabase

05Multi-source scraping templates for HTML, REST APIs, and RSS feeds

06172,009 GitHub stars

Use Cases

01Tracking product prices and sending alerts for significant price drops

02Aggregating and classifying technical news or GitHub releases by sentiment and topic

03Monitoring job boards to find and score roles matching a specific resume