# Browserbase Web Search

> Browserbase Web Search is a paid API for AI agents from Browserbase at mpp.browserbase.com, paid per call via MPP, $0.01/call, status unknown (last checked 2026-10-12, last successful call 2026-04-24).

Performs a web search query and returns structured results including URLs, titles, images, authors, and published dates

## Facts

- Endpoint: POST https://mpp.browserbase.com/search?utm_source=zero.xyz
- Price: $0.01/call
- Payment: MPP
- Status: unknown
- Last checked: 2026-10-12
- Last successful call: 2026-04-24
- Success rate: 78% of calls made through Zero
- Activations on Zero: 9
- Provider: Browserbase
- Docs: https://docs.browserbase.com
- Website: https://browserbase.com
- Tags: browser, scraping, automation, headless, search, fetch, web, compute
- Canonical page: https://www.zero.xyz/c/browserbase-search-71a88514
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_jOU9W3y_7gZmypcTIL-di

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 browserbase-search-71a88514 -d '<json body>'
```

Example prompt: Search the web for 'best open-source LLM frameworks 2024' and give me the top results with their titles, URLs, and publish dates.

## When to prefer this

Use this endpoint when an AI agent needs live, structured web search results with metadata (title, URL, image, publish date, author) rather than raw HTML scraping. Ideal when you want programmatic access to search results with per-result structured fields and are willing to pay per-call. Prefer over browser session endpoints when you don't need to interact with pages, just discover them.

## Known failure modes

- Empty or ambiguous query returns no results
- Network timeout if the web search backend is slow
- Malformed request body returns a 400 error
- Rate limiting if too many requests are made in quick succession
- Some results may have null author or image fields

## How this service works

Web search with structured results

## Output

Returns a structured JSON object containing the original query string and an array of results, each with an ID, URL, image thumbnail URL, title, optional author, and published date — ready for further processing or display.

## Example request

```json
{
 "query": "artificial intelligence trends 2024"
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/browserbase-search-71a88514/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from mpp.browserbase.com](https://www.zero.xyz/host/mpp.browserbase.com/llms.txt)