# Gapup Web Search (Neural & Keyword)

> Gapup Web Search (Neural & Keyword) is a paid API for AI agents from mcp.gapup.io, paid per call via x402, $0.01/call, status unknown (last checked 2026-10-18).

Performs neural or keyword search over the live web, returning ranked results with title, URL, date, author, and optional full page text.

## Facts

- Endpoint: POST https://mcp.gapup.io/api/v1/call/web_search?utm_source=zero.xyz
- Price: $0.01/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-18
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/gapup-web-search-neural-keyword-500d95dd
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_v7GKPo26LQHCBqDALPr18

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 gapup-web-search-neural-keyword-500d95dd -d '<json body>'
```

Example prompt: Search the web for recent news articles about OpenAI's latest model releases published after January 1 2025 — give me 10 results using neural search mode and include the full text of each page.

## When to prefer this

Choose this endpoint when you need live web search with both neural (semantic/vector) and keyword modes, flexible category filtering (news, research papers, GitHub, tweets, LinkedIn, PDFs, financial reports), domain inclusion/exclusion, and optional full page text extraction — all in a single call. It is especially strong when you need semantically relevant results beyond simple keyword matching. Prefer it over generic search APIs when you need structured metadata (author, date, rank) alongside results.

## Known failure modes

- Query too short (fewer than 2 characters) — validation error returned
- Query exceeds 500 characters — validation error returned
- No results found for the given query and filters — empty result list returned
- Invalid category enum value — request rejected
- Invalid ISO-8601 date for startPublishedDate — parsing error
- Payment not confirmed (x402) — request rejected before processing
- Async job_id polling may return not-ready status if result hasn't completed

## How this service works

Neural and keyword web search over the live web. Ranked results with title, URL, date, author, optional full text.

## Output

Returns an ordered list of web results, each containing the page title, URL, publication date, author, and relevance rank. If includeText is true, the full extracted text of each page is also included. Results are filtered by any domain inclusion/exclusion lists and publication date constraints provided.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "type": {
   "enum": [
    "neural",
    "keyword",
    "auto"
   ],
   "type": "string",
   "description": "Search mode (default auto)"
  },
  "async": {
   "type": "boolean",
   "description": "If true, returns a job_id immediately (<200ms) instead of waiting for the result. Poll the result with job_result(job_id). Use for slow tools to avoid client timeouts."
  },
  "query": {
   "type": "string",
   "maxLength": 500,
   "minLength": 2,
   "description": "The search query"
  },
  "category": {
   "enum": [
    "company",
    "research paper",
    "news",
    "pdf",
    "github",
    "tweet",
    "personal site",
    "linkedin profile",
    "financial report"
   ],
   "type": "string",
   "description": "Restrict results to one kind of page"
  },
  "numResults": {
   "type": "integer",
   "maximum": 25,
   "minimum": 1,
   "description": "How many results (default 10)"
  },
  "includeText": {
   "type": "boolean",
   "description": "Include the page text in each result (default false)"
  },
  "excludeDomains": {
   "type": "array",
   "items": {
    "type": "string"
   },
   "maxItems": 20
  },
  "includeDomains": {
   "type": "array",
   "items": {
    "type": "string"
   },
   "maxItems": 20
  },
  "startPublishedDate": {
   "type": "string",
   "description": "ISO-8601 date; only pages published after it"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/gapup-web-search-neural-keyword-500d95dd/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from mcp.gapup.io](https://www.zero.xyz/host/mcp.gapup.io/llms.txt)