crw-extract

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

crw-extract — typed JSON from pages

crw-extract — 从网页获取类型化JSON

When to use

适用场景

You need a structured JSON object from a page, not prose.
Step 6 in the crw ladder. First scrape succeeds (step 2), but you need machine-readable fields. If the source is a local PDF, use crw-parse (step 5) with
```
formats:["json"]
```
instead.
Schema-driven = deterministic output shape. No schema = exploratory; use a
```
prompt
```
to describe what you want.

你需要从页面获取结构化JSON对象，而非散文式内容。
这是crw工作流阶梯的第6步。当第一步抓取成功（第2步）但你需要机器可读字段时使用。如果数据源是本地PDF，请改用crw-parse（第5步）并设置
```
formats:["json"]
```
。
基于Schema驱动 = 输出形状确定。无Schema = 探索式提取；使用
```
prompt
```
描述你想要的内容。

How extraction works in crw

crw中的提取原理

crw has no
/v1/extract
endpoint (Firecrawl's dedicated extract API). Instead, extraction runs in two ways, both backed by the same LLM pipeline:

Path	When to use	Sync?
Per-page: `formats:["json"]` + `jsonSchema` on scrape	Single URL, or inline during a crawl	Synchronous
Async multi-URL: `POST /v2/extract` → poll `GET /v2/extract/{id}`	Many URLs, fire-and-forget	Async

/v2/extract

is marked deprecated in the server (it recommends

/v2/scrape

with

formats:["json"]

), but it works and is useful for multi-URL batches.

Requires a server-side LLM. Set

[extraction.llm]

in the server config (provider, api_key, model). Without it, requests return an error (HTTP 4xx) — e.g. 422 "no LLM configured". Use

crw setup

to configure the LLM for the CLI.

crw没有专门的
/v1/extract
端点（Firecrawl的专用提取API）。相反，提取通过两种方式运行，均由同一LLM管道提供支持：

方式	适用场景	是否同步
单页提取：抓取时设置 `formats:["json"]` + `jsonSchema`	单个URL，或在抓取过程中实时提取	同步
异步多URL提取： `POST /v2/extract` → 轮询 `GET /v2/extract/{id}`	多个URL，提交后无需等待	异步

/v2/extract

在服务器中标记为已弃用（推荐使用带有

formats:["json"]

的

/v2/scrape

），但它仍可正常工作，且对批量多URL提取很有用。

需要服务器端LLM。在服务器配置中设置

[extraction.llm]

（提供商、api_key、模型）。如果未配置，请求将返回错误（HTTP 4xx）——例如422错误“no LLM configured”。使用

crw setup

为CLI配置LLM。

Quick start

快速开始

CLI — per-page extraction via

--extract

bash

undefined

CLI — 通过

--extract

进行单页提取：

bash

undefined

Inline schema

内联Schema

crw scrape "https://example.com/product"
--extract '{"type":"object","properties":{"price":{"type":"number"},"inStock":{"type":"boolean"}}}'

Schema from file

从文件加载Schema

crw scrape "https://example.com/job" --extract @schema.json -o result.json


**MCP** — pass `formats:["json"]` with `jsonSchema` on a scrape:

crw_scrape( url="https://example.com/product", formats=["json"], extract={"schema": {"type":"object","properties":{"price":{"type":"number"}}}} )


Note: the MCP `crw_scrape` accepts `extract.schema` (Firecrawl style). The
REST API also accepts `jsonSchema` as a top-level alias.

**REST** — per-page (synchronous):

```bash
curl -X POST "$CRW_API_URL/v1/scrape" \
  -H "Authorization: Bearer $CRW_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://example.com/product",
    "formats": ["json"],
    "jsonSchema": {
      "type": "object",
      "properties": {
        "price": {"type": "number"},
        "inStock": {"type": "boolean"}
      }
    }
  }'

REST — async multi-URL (deprecated endpoint, still functional):

bash

undefined

crw scrape "https://example.com/job" --extract @schema.json -o result.json


**MCP** — 在抓取时传入`formats:["json"]`和`jsonSchema`：

crw_scrape( url="https://example.com/product", formats=["json"], extract={"schema": {"type":"object","properties":{"price":{"type":"number"}}}} )


注意：MCP的`crw_scrape`接受`extract.schema`（Firecrawl风格）。REST API也接受`jsonSchema`作为顶级别名。

**REST** — 单页提取（同步）：

```bash
curl -X POST "$CRW_API_URL/v1/scrape" \
  -H "Authorization: Bearer $CRW_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://example.com/product",
    "formats": ["json"],
    "jsonSchema": {
      "type": "object",
      "properties": {
        "price": {"type": "number"},
        "inStock": {"type": "boolean"}
      }
    }
  }'

REST — 异步多URL提取（已弃用端点，仍可使用）：

bash

undefined

Start job

启动任务

curl -X POST "$CRW_API_URL/v2/extract"
-H "Authorization: Bearer $CRW_API_KEY"
-H "Content-Type: application/json"
-d '{ "urls": ["https://example.com/p1", "https://example.com/p2"], "schema": {"type":"object","properties":{"price":{"type":"number"}}} }'

→ {"success":true,"id":"<uuid>","warnings":["...use /v2/scrape..."], ...}

Poll until completed

轮询直到完成

curl "$CRW_API_URL/v2/extract/<uuid>" -H "Authorization: Bearer $CRW_API_KEY"

→ {"success":true,"status":"completed|scraping|failed","data":{...}}

undefined

undefined

Options

选项

Need	CLI	MCP / REST
JSON schema	`--extract '<schema>'` or `@file.json`	`jsonSchema` / `extract.schema`
Free-text prompt (no schema)	—	`prompt` on `/v2/extract`
Save output	`-o FILE`	write the response yourself
Multi-URL async	not available	`POST /v2/extract` with `urls:[...]`
LLM override	`--llm-provider` , `--llm-key` , `--llm-model`	server config only

需求	CLI	MCP / REST
JSON Schema	`--extract '<schema>'` 或 `@file.json`	`jsonSchema` / `extract.schema`
自由文本提示（无Schema）	—	在 `/v2/extract` 中使用 `prompt`
保存输出	`-o FILE`	自行处理响应并写入
多URL异步提取	不支持	使用带有 `urls:[...]` 的 `POST /v2/extract`
LLM覆盖配置	`--llm-provider` , `--llm-key` , `--llm-model`	仅支持服务器配置

Tips

提示技巧

Schema = deterministic, prompt = exploratory. A schema pins the output shape; a free-text
```
prompt
```
is useful for exploration but less reliable. Start with a schema when you know the fields you want.
Don't over-specify. Narrow schemas ("give me exactly these three fields") extract more reliably than wide ones with fifty optional fields.
Crawl + extract in one pass.
```
crw_crawl
```
accepts a
```
jsonSchema
```
parameter — each page in the crawl gets extracted against the schema, saving a second round-trip.
Check
data.json
in the scrape response. Per-page extraction lands in
```
data.json
```
, not
```
data.markdown
```
. The
```
markdown
```
field is also populated for reference.

Schema确保确定性，提示用于探索。Schema固定输出形状；自由文本
```
prompt
```
适用于探索但可靠性较低。当你明确需要哪些字段时，优先使用Schema。
不要过度指定Schema。范围狭窄的Schema（“只给我这三个字段”）比包含五十个可选字段的宽泛Schema提取更可靠。
一次完成抓取+提取。
```
crw_crawl
```
接受
```
jsonSchema
```
参数——抓取中的每个页面都会根据Schema进行提取，节省一次往返请求。
检查抓取响应中的
data.json
。单页提取结果会放在
```
data.json
```
中，而非
```
data.markdown
```
。
```
markdown
```
字段也会填充，供参考使用。

crw-extract

Original

Translation

crw-extract — typed JSON from pages

crw-extract — 从网页获取类型化JSON

When to use

适用场景

How extraction works in crw

crw中的提取原理

Quick start

快速开始

Inline schema

内联Schema

Schema from file

从文件加载Schema

Start job

启动任务

→ {"success":true,"id":"<uuid>","warnings":["...use /v2/scrape..."], ...}

→ {"success":true,"id":"<uuid>","warnings":["...use /v2/scrape..."], ...}

Poll until completed

轮询直到完成

→ {"success":true,"status":"completed|scraping|failed","data":{...}}

→ {"success":true,"status":"completed|scraping|failed","data":{...}}

Options

选项

Tips

提示技巧

See also

相关链接