---
title: "Search the web and optionally scrape results"
method: POST
path: "/v1/search"
tags: ["search"]
---

# Search the web and optionally scrape results

`POST /v1/search`

## Request

- Base URL: `https://api.fastcrw.com`
- URL: `https://api.fastcrw.com/v1/search`
- Auth: one of:
  - HTTP bearer
  - none

## Request body

- SearchRequest
  - `query` string, required
  - `limit` integer
  - `lang` string
  - `tbs` 'qdr:h' | 'qdr:d' | 'qdr:w' | 'qdr:m' | 'qdr:y'
  - `sources` string[]
  - `categories` string[]
  - `scrapeOptions` SearchScrapeOptions
    - `formats` string[]
    - `onlyMainContent` boolean
    - `timeout` integer: Per-result scrape budget in ms. Default 15000. Search awaits every result, so this bounds the whole response; raise it for heavy JS pages.

## Response `200`

Search results

- SearchResponse
  - `success` boolean, required
  - `data` union, required: Search results. On the hosted API (https://api.fastcrw.com) the results ARE the value of `data`. A self-hosted engine wraps them one level deeper as `data.results`, and carries `answer`/`citations`/`llmUsage`/`warnings` inside `data` instead of at the top level. Clients that must work against both should read `data.results` when present and fall back to `data`.
    - SearchResult[]: Returned when `sources` is not set.
      - `url` string, uri, required
      - `title` string, required
      - `snippet` string: Result blurb (LLM-ready summary line)
      - `description` string: Alias of snippet (same value, kept for Firecrawl response-shape parity)
      - `position` integer
      - `score` number
      - `category` string: Search backend result category (e.g. 'general')
      - `markdown` string: Present when scrapeOptions.formats includes 'markdown'
    - object: Returned when `sources` is set.
      - `web` SearchResult[]
        - `url` string, uri, required
        - `title` string, required
        - `snippet` string: Result blurb (LLM-ready summary line)
        - `description` string: Alias of snippet (same value, kept for Firecrawl response-shape parity)
        - `position` integer
        - `score` number
        - `category` string: Search backend result category (e.g. 'general')
        - `markdown` string: Present when scrapeOptions.formats includes 'markdown'
      - `news` SearchResult[]
        - `url` string, uri, required
        - `title` string, required
        - `snippet` string: Result blurb (LLM-ready summary line)
        - `description` string: Alias of snippet (same value, kept for Firecrawl response-shape parity)
        - `position` integer
        - `score` number
        - `category` string: Search backend result category (e.g. 'general')
        - `markdown` string: Present when scrapeOptions.formats includes 'markdown'
      - `images` object[]
    - object: The self-hosted engine nests the flat or grouped results under `results`.
      - `results` unknown, required
      - `answer` string
      - `citations` object[]
      - `llmUsage` object
      - `warnings` string[]
  - `answer` string: Hosted API only — synthesized answer over the top results, present when `answer: true` was set. A self-hosted engine returns this inside `data`.
  - `citations` object[]: Hosted API only — sources for `answer`, in citation order. A self-hosted engine returns this inside `data`.
  - `llmUsage` object: Hosted API only — token usage for the answer synthesis. A self-hosted engine returns this inside `data`.
  - `warnings` string[]: Hosted API only — soft-failure notices (e.g. an answer leg that was rate-limited). A self-hosted engine returns this inside `data`.

## Other responses

- `400`: Invalid request body
- `401`: Missing or invalid bearer token (hosted only)
- `429`: Rate limit exceeded
- `503`: Search is disabled in self-hosted config (error: search_disabled)

---

[API](https://skmtc.dev/fastcrw/apis/crw-rest-api.md) · [All operations](https://skmtc.dev/fastcrw/apis/crw-rest-api/llms.txt) · [OpenAPI document](https://skmtc.dev/fastcrw/apis/crw-rest-api/revisions/9bc1842aa048?raw)