# Venice AI Web Scraper

> Venice AI Web Scraper is a paid API for AI agents from api.venice.ai, paid per call via x402, $10/call, status unknown (last checked 2026-10-11).

Scrapes a public web page and returns its full content converted to Markdown format

## Facts

- Endpoint: POST https://api.venice.ai/api/v1/augment/scrape?utm_source=zero.xyz
- Price: $10/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-11
- Activations on Zero: 0
- Tags: x402, ai_ml
- Canonical page: https://www.zero.xyz/c/venice-ai-web-scraper-22a9ca1b
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_g1s-qNZcJS5_vhO_9rdgf

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 venice-ai-web-scraper-22a9ca1b -d '<json body>'
```

Example prompt: Can you scrape the page at https://example.com/article and give me its content as markdown?

## When to prefer this

Use this endpoint when you need to retrieve and read the textual content of a public web page and want it delivered in clean Markdown format rather than raw HTML. It is particularly useful for AI agents that need to ingest article text, documentation, or product pages for summarization, question answering, or data extraction. Choose this over generic HTTP fetch when you need automatic HTML-to-Markdown conversion. Be aware that certain sites like Twitter/X and Reddit are explicitly blocked.

## Known failure modes

- URL belongs to a blocked site (e.g. X/Twitter, Reddit) — returns an immediate rejection error
- Malformed or non-URI URL input — returns validation error
- Target page requires authentication or is behind a paywall — may return incomplete or access-denied content
- Insufficient x402 wallet balance — returns 402 Payment Required with top-up instructions
- Network timeout or unreachable target URL — returns error response

## How this service works

Scrape a web page and return its content as markdown. Supports most public web pages; some sites (e.g. X/Twitter, Reddit) that block automated access are rejected immediately.

**Authentication:** This endpoint accepts either a Bearer API key or an `X-Sign-In-With-X` header for x402 wallet-based authentication. When using x402, a `402 Payment Required` response indicates insufficient balance and includes top-up instructions.

## Output

Returns a JSON object containing the original URL, the format field ('markdown'), and the full page content converted to Markdown text, suitable for reading, summarizing, or further processing.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "url"
 ],
 "properties": {
  "url": {
   "type": "string",
   "format": "uri",
   "example": "https://example.com",
   "description": "The URL to scrape"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "url",
  "content",
  "format"
 ],
 "properties": {
  "url": {
   "type": "string",
   "example": "https://example.com",
   "description": "The URL that was scraped"
  },
  "format": {
   "enum": [
    "markdown"
   ],
   "type": "string",
   "description": "The format of the scraped content"
  },
  "content": {
   "type": "string",
   "description": "The scraped content in markdown format"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/venice-ai-web-scraper-22a9ca1b/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from api.venice.ai](https://www.zero.xyz/host/api.venice.ai/llms.txt)