# glim.sh Web Search API

> glim.sh Web Search API is a paid API for AI agents from surf.cascade.fyi, paid per call via x402, $0.01/call, status healthy (last checked 2026-09-15, last successful call 2026-08-28).

Searches the live web (including Twitter, Reddit, GitHub, and open web) and returns ranked results with titles, snippets, authors, and publish dates — no API keys required, pay-per-call.

## Facts

- Endpoint: POST https://surf.cascade.fyi/api/v1/web/search
- Price: $0.01/call
- Payment: x402
- Status: healthy
- Last checked: 2026-09-15
- Last successful call: 2026-08-28
- Success rate: 50% of calls made through Zero
- Rating: 4.0 / 5 from 1 review
- Activations on Zero: 2
- Tags: x402
- Canonical page: https://www.zero.xyz/c/glim-sh-web-search-api-132c72b4
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_lOwKA9RMF9ho3b7zqxZQR

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-search-api-132c72b4 -d '<json body>'
```

Example prompt: Search the live web for what people on Reddit and Twitter are saying about OpenAI's latest model release in the past 7 days, excluding pinterest.com and quora.com from the results.

## When to prefer this

Choose this endpoint when an AI agent needs live, up-to-date information from the open web, Twitter, Reddit, or GitHub without managing API keys or scraping infrastructure. It is especially useful for social listening, recent news retrieval, research queries, and monitoring what is being said about a topic right now. Prefer it over static knowledge bases when freshness matters.

## Known failure modes

- Query returns zero results if the topic is too niche or published_within_days is too restrictive
- Included domains may have no indexed results for the query, returning an empty set
- Overly broad queries may return low-relevance results with poor scores
- Payment failure (402) if USDC balance is insufficient or x402 handshake fails
- Rate limiting or timeout if the underlying live-data sources are slow to respond

## How this service works

Web search powered by Exa

## Output

A JSON array of ranked web results, each containing a URL, relevance score (0–1), page title, author name, text snippet, and published date. The count of results is also returned.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "query": {
   "type": "string",
   "maxLength": 1000,
   "minLength": 1,
   "description": "Search query"
  },
  "exclude_domains": {
   "type": "array",
   "items": {
    "type": "string"
   },
   "description": "Exclude results from these domains. Useful for filtering noisy aggregators or SEO farms when you've seen them dominate results (e.g. ['pinterest.com', 'quora.com'])."
  },
  "include_domains": {
   "type": "array",
   "items": {
    "type": "string"
   },
   "description": "Restrict results to these domains (e.g. ['arxiv.org'])"
  },
  "published_within_days": {
   "type": "integer",
   "maximum": 3650,
   "minimum": 1,
   "description": "Restrict to results published within the last N days. Skip this for broad queries - it excludes pages without publish-date metadata. Common values: 1, 7, 30, 365."
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "count": 1,
  "results": [
   {
    "url": "https://example.com",
    "score": 0.95,
    "title": "Example",
    "author": "Author Name",
    "snippet": "...",
    "published_date": "2026-01-15"
   }
  ]
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/glim-sh-web-search-api-132c72b4/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)