# GEDX402 Browser Safe-Extract

> GEDX402 Browser Safe-Extract is a paid API for AI agents from browser.gedx402.com, paid per call via x402, $0.24/call, status unknown (last checked 2026-09-16).

Loads a URL in a headless browser, performs safety checks, and optionally extracts structured content (article, product, or full) plus Markdown — paid via USDC on-chain with no API key required.

## Facts

- Endpoint: GET https://browser.gedx402.com/v1/browser/outcome/safe-extract
- Price: $0.24/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-16
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/gedx402-browser-safe-extract-56678414
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_att74K9VT_DVXQWZ12G9r

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 gedx402-browser-safe-extract-56678414
```

Example prompt: Check if https://www.bbc.com/news/world-us-canada-12345678 is safe and extract the article text with title and author using the article recipe.

## When to prefer this

Choose this endpoint when you need both a safety check and structured content extraction from a URL in a single call, especially when you have no API key available and want to pay per-use with USDC. Ideal for AI agents that need to safely ingest arbitrary web URLs into article, product, or full-page structured data without managing credentials.

## Known failure modes

- URL is unreachable or returns non-200 status — safe flag may be false or extraction fields empty
- Unsafe page detected — safe:false returned, content may be withheld
- Invalid recipe enum value — request rejected or falls back to Markdown-only
- Payment failure on-chain — 402 response, no content returned
- URL is not HTTPS — rejected per input schema requirement
- Page renders slowly or has bot protection — incomplete or empty extraction

## How this service works

Sentinel-like safe extract — one x402 payment runs Llama Guard on page Markdown, then returns { safe, categories?, markdown?, ... } or blocks with reason. POST { url, recipe?: article|product|full } on browser.gedx402.com. x402 settlement $0.24. Omit recipe for guard + Markdown only. Discover: GET /v1/models?bundle=browser.

## Output

A JSON object containing: a 'safe' boolean indicating whether the page passed safety checks, the recipe used, full Markdown rendering of the page, structured metadata (title, author, etc.), and categories — all extracted by a headless browser render.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "url": {
   "type": "string"
  },
  "recipe": {
   "enum": [
    "article",
    "product",
    "full"
   ],
   "type": "string"
  },
  "options": {
   "type": "object",
   "additionalProperties": true
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/gedx402-browser-safe-extract-56678414/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from browser.gedx402.com](https://www.zero.xyz/host/browser.gedx402.com/llms.txt)