How to use AnyCrawler Skills with your AI agent

Hey Product Hunt community,

We just launched AnyCrawler, and I want to share a simple way to use it directly inside AI agent workflows: AnyCrawler Skills.

AnyCrawler Skills let an AI agent read webpages, capture screenshots, and run public search through AnyCrawler’s API without you manually wiring crawler logic into every prompt or script.

There are two skills available:

• $anycrawler-read
  For crawling webpages, rendering pages, extracting readable markdown, and taking screenshots.

• $anycrawler-search
  For searching web pages, images, news, videos, and scholar results.

Setup

1. Create an AnyCrawler account: https://app.anycrawler.com

2. Get an API key from the dashboard.

3. Install the skills from GitHub: https://github.com/AnyCrawler-com/AnyCrawler-Skill

4. Set your API key as an environment variable:

bash

export ANYCRAWLER_API_KEY="sk-your-key"

On Windows PowerShell:

powershell

$env:ANYCRAWLER_API_KEY = "sk-your-key"

Then restart your AI agent session so it can discover the new skills.

Example prompts

Use $anycrawler-read to crawl https://example.com with render and save markdown.

Use $anycrawler-read to take a screenshot of https://example.com and download the PNG.

Use $anycrawler-search to search recent news about AI web crawling.

The goal is to make web crawling and search feel like a native agent capability: give the agent a task, and let the skill handle the API calls, response format, and reusable workflow.

This is especially useful for research agents, SEO tools, market monitoring, lead enrichment, content workflows, and any app that needs structured web data.

Repo: https://github.com/AnyCrawler-com/AnyCrawler-Skill
API: https://api.anycrawler.com
Dashboard: https://app.anycrawler.com

Would love feedback from builders using AI agents in real workflows.