# Agent Kiosk Web Extract

> Agent Kiosk Web Extract is a paid API for AI agents from agent-kiosk.vercel.app, paid per call via x402, $0.01/call, status unknown (last checked 2026-10-17).

Fetches a URL and returns its content as clean, readable markdown text, stripping ads, navigation, and boilerplate.

## Facts

- Endpoint: POST https://agent-kiosk.vercel.app/v1/web-extract?utm_source=zero.xyz
- Price: $0.01/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-17
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/agent-kiosk-web-extract-893bf747
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_1S6NAv5WDd8rWJOZWX6R3

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 agent-kiosk-web-extract-893bf747 -d '<json body>'
```

Example prompt: Can you fetch the content of https://techcrunch.com/2024/01/15/openai-announcement/ and give me the readable article text as markdown, up to 10000 characters?

## When to prefer this

Choose this endpoint when you need to retrieve and read the human-readable content of a specific known URL and want it returned as clean markdown, especially for one-off page reads in an agent workflow. It is pay-per-call at $0.01 USDC via x402 micropayment, making it cost-effective for occasional fetches without a subscription. Prefer it over browser automation tools when you just need the text body, not screenshots or JavaScript interaction.

## Known failure modes

- URL is unreachable or returns a non-200 HTTP status — ok: false with error detail
- URL does not start with http:// or https:// — validation error returned
- Page is behind a login wall or CAPTCHA — content may be incomplete or empty
- max_length out of allowed range (500-25000) — rejected with validation error
- Rate limiting or site blocking the scraper — partial or empty content returned

## How this service works

Clean webpage-to-Markdown extractor with automatic Prompt Injection trap scanning for AI agents.

## Output

Returns a JSON object with an 'ok' boolean and a 'data' object containing the extracted page content as clean markdown text, stripped of ads, navigation menus, and boilerplate HTML. Also includes a 'payment_mode' field indicating how the x402 micropayment was processed.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "url": {
   "type": "string",
   "description": "Target webpage URL (https://...)"
  },
  "max_length": {
   "type": "number"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "url": "https://example.com",
  "title": "Example Domain",
  "llm_markdown": "# Example Domain\nThis domain is for use in illustrative examples...",
  "prompt_injection_scan": {
   "safe": true,
   "risk_level": "LOW"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/agent-kiosk-web-extract-893bf747/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from agent-kiosk.vercel.app](https://www.zero.xyz/host/agent-kiosk.vercel.app/llms.txt)