Scrapegraph Search

ScrapeGraph AI Web Scraping Skill (scrapegraph-mcp)

This skill provides advanced AI-powered web scraping, structured extraction, search, async crawling, schema synthesis, and monitoring via the ScrapeGraph MCP server (https://sgai-mcp-main.onrender.com).

Available Tools & Signatures

1. scrape

Scrapes web content with specified output formats.

Arguments:

• url (string, required): Web page URL to scrape.
• output_format (string, optional): "markdown", "html", "screenshot", "branding", "links", "images", "summary".

2. extract

Extracts structured data from a target website using an AI user prompt and schema.

Arguments:

• website_url (string, required): Target URL.
• user_prompt (string, required): Natural language extraction instruction.
• output_schema (object, optional): Desired JSON schema for extracted output.

3. search

Executes search query and extracts web content.

Arguments:

• query (string, required): Search query.
• num_results (integer, optional): Number of results (3 to 20).
• country_search (string, optional): Geographic bias.

4. crawl_start / crawl_get_status / crawl_stop

Initiates async multi-page crawling and retrieves status results.

5. schema

Generates or augments JSON schema from a natural language prompt.

Usage Guidelines & Best Practices

1. Use scrape with output_format: "markdown" for clean text extraction.
2. Use extract when specific JSON structures or attributes (prices, tables, specs) are requested.
3. Requires SGAI_API_KEY set in the environment or request header X-API-Key.