# ScrapeGraphAI LLM-Powered Web Extraction

> ScrapeGraphAI LLM-Powered Web Extraction is a paid API for AI agents from x402.orth.sh, paid per call via x402, $0.025/call, status unknown (last checked 2026-09-15).

Extracts structured or free-form data from a URL, raw HTML, or markdown using an LLM-driven pipeline with a natural language prompt and optional JSON schema for structured output.

## Facts

- Endpoint: POST https://x402.orth.sh/scrapegraphai/api/extract
- Price: $0.025/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-15
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/scrapegraphai-llm-powered-web-extraction-f6617614
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_MumS_E57n9Qi-lFBHOdOe

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability scrapegraphai-llm-powered-web-extraction-f6617614 -d '<json body>'
```

Example prompt: Go to https://news.ycombinator.com and extract a list of the top 10 story titles and their URLs — return it as structured JSON with fields 'title' and 'url' for each item, using reader mode to clean up the page first.

## When to prefer this

Choose this endpoint when you need AI-driven, prompt-guided extraction from a specific URL or raw content — especially when the output needs to conform to a JSON schema. Prefer this over generic scrapers when the data is semi-structured or requires interpretation (e.g. extracting sentiment, summarizing paragraphs, or pulling fields embedded in prose). It is ideal when you already have the HTML or markdown and just need structured extraction without re-fetching.

## Known failure modes

- URL is unreachable or returns a non-200 status — extraction fails with a fetch error
- Prompt is too vague and LLM returns incomplete or hallucinated fields
- Source HTML is too large or complex and gets truncated
- Schema mismatch: LLM cannot reliably populate all required schema fields from the content
- Stealth/fetch mode insufficient for JavaScript-heavy SPAs — returns empty or partial content
- Rate limiting or blocking by the target site — returns empty fetch result

## How this service works

LLM-driven extraction from a URL, HTML, or markdown input with a prompt and optional JSON schema.

## Output

Returns extracted data shaped according to the optional JSON schema provided, or as free-form LLM-generated output if no schema is given. Fields are populated based on what the LLM interprets from the source content (URL-fetched, raw HTML, or markdown) in response to the natural language prompt.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "url": {
   "type": "string",
   "description": "URL to extract from"
  },
  "html": {
   "type": "string",
   "description": "Raw HTML to extract from"
  },
  "mode": {
   "type": "string",
   "description": "Content mode: normal, reader, or prune"
  },
  "prompt": {
   "type": "string",
   "description": "Extraction prompt (1-10000 chars)"
  },
  "schema": {
   "type": "object",
   "description": "JSON schema for structured output"
  },
  "markdown": {
   "type": "string",
   "description": "Markdown to extract from"
  },
  "contentType": {
   "type": "string",
   "description": "Force content type"
  },
  "fetchConfig": {
   "type": "object",
   "description": "Fetch options: mode, stealth, timeout, wait, headers, cookies, country, scrolls"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/scrapegraphai-llm-powered-web-extraction-f6617614/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from x402.orth.sh](https://www.zero.xyz/host/x402.orth.sh/llms.txt)