# glim.sh Web Fetch (surf.cascade.fyi)

> glim.sh Web Fetch (surf.cascade.fyi) is a paid API for AI agents from surf.cascade.fyi, paid per call via x402, $0.002/call, status unknown (last checked 2026-09-13).

Fetches and returns the live content of any URL (web pages, Twitter, Reddit, GitHub, etc.) in markdown, HTML, or JSON format — no API keys required.

## Facts

- Endpoint: POST https://surf.cascade.fyi/api/v1/web/fetch
- Price: $0.002/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-13
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/glim-sh-web-fetch-live-web-content-retrieval-234b68c0
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_GuT19gu2vEOyL76Milfuz

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 glim-sh-web-fetch-live-web-content-retrieval-234b68c0 -d '<json body>'
```

Example prompt: Fetch the live content of https://www.reddit.com/r/MachineLearning/top/ and return it as markdown so I can see what's trending right now.

## When to prefer this

Choose this endpoint when an AI agent needs live, real-time content from any public URL — including social platforms like Twitter and Reddit — without managing API keys, rate limits, or a dedicated scraping infrastructure. Ideal for pay-per-call, low-overhead web access at $0.002 per call.

## Known failure modes

- URL is unreachable or returns a non-200 status — status field reflects the HTTP error code
- Paywalled or login-required pages may return incomplete or blocked content
- Payment not included or insufficient — returns HTTP 402 requiring $0.002 USDC payment
- Invalid format enum value — request rejected with validation error
- URL field missing — request rejected

## How this service works

One remote endpoint gives your agent live data from Twitter, Reddit, the open web, GitHub and more. No API keys, no scraping stack - just connect the URL and go.

## Output

Returns a JSON object containing the original URL, HTTP status code, the fetched page content (as markdown, HTML, or JSON depending on the requested format), and the total character count of the content.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "url": {
   "type": "string",
   "maxLength": 2048,
   "description": "URL to fetch"
  },
  "format": {
   "enum": [
    "markdown",
    "html",
    "json"
   ],
   "type": "string",
   "default": "markdown",
   "description": "Output format. 'markdown' (default) clean article text; 'html' raw cleaned HTML; 'json' the structured SSR blob (TikTok, Pinterest, etc.) instead of article text."
  },
  "selector": {
   "type": "string",
   "maxLength": 500,
   "description": "CSS selector to scope extraction"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "url": "https://example.com",
  "status": 200,
  "content": [
   "# Page Title\nContent..."
  ],
  "total_chars": 1234
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/glim-sh-web-fetch-live-web-content-retrieval-234b68c0/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from surf.cascade.fyi](https://www.zero.xyz/host/surf.cascade.fyi/llms.txt)