pdfco

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

PDF.co

All-in-one PDF processing API. Convert, extract, merge, split, compress PDFs and more. Supports OCR for scanned documents.

Official docs: https://docs.pdf.co/

一站式PDF处理API，支持PDF转换、提取、合并、拆分、压缩等功能，可对扫描文档提供OCR支持。

官方文档：https://docs.pdf.co/

When to Use

适用场景

Use this skill when you need to:

Extract text from PDF files (with OCR support)
Convert PDF to CSV, JSON, or other formats
Merge multiple PDFs into one
Split PDF into multiple files
Compress PDF to reduce file size
Convert HTML/URL to PDF
Parse invoices and documents with AI

当你需要完成以下操作时可以使用这个工具：

从PDF文件提取文本（支持OCR）
将PDF转换为CSV、JSON或其他格式
合并多个PDF为一个文件
将PDF拆分为多个文件
压缩PDF减小文件体积
将HTML/URL转换为PDF
用AI解析发票和文档

Prerequisites

前提条件

Create an account at https://pdf.co/
Get your API key from https://app.pdf.co/

Set environment variable:

bash

export PDFCO_API_KEY="your-email@example.com_your-api-key"

Important: When using
$VAR
in a command that pipes to another command, wrap the command containing
$VAR
in
bash -c '...'
. Due to a Claude Code bug, environment variables are silently cleared when pipes are used directly.
bash
bash -c 'curl -s "https://api.example.com" -H "Authorization: Bearer $API_KEY"'

在https://pdf.co/ 创建账号
从https://app.pdf.co/ 获取你的API密钥

设置环境变量：

bash

export PDFCO_API_KEY="your-email@example.com_your-api-key"

重要提示： 在需要管道传输到其他命令的命令中使用
$VAR
时，请将包含
$VAR
的命令包裹在
bash -c '...'
中。由于Claude Code的bug，直接使用管道时环境变量会被静默清空。
bash
bash -c 'curl -s "https://api.example.com" -H "Authorization: Bearer $API_KEY"'

How to Use

使用指南

1. PDF to Text

1. PDF转文本

Extract text from PDF with OCR support:

Write to

/tmp/request.json

json

{
  "url": "https://pdfco-test-files.s3.us-west-2.amazonaws.com/pdf-to-text/sample.pdf",
  "inline": true
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/pdf/convert/to/text" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

With specific pages (1-indexed):

Write to

/tmp/request.json

json

{
  "url": "https://pdfco-test-files.s3.us-west-2.amazonaws.com/pdf-to-text/sample.pdf",
  "pages": "1-3",
  "inline": true
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/pdf/convert/to/text" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

支持OCR从PDF中提取文本：

写入到

/tmp/request.json

：

json

{
  "url": "https://pdfco-test-files.s3.us-west-2.amazonaws.com/pdf-to-text/sample.pdf",
  "inline": true
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/pdf/convert/to/text" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

指定页码（从1开始计数）：

写入到

/tmp/request.json

：

json

{
  "url": "https://pdfco-test-files.s3.us-west-2.amazonaws.com/pdf-to-text/sample.pdf",
  "pages": "1-3",
  "inline": true
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/pdf/convert/to/text" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

2. PDF to CSV

2. PDF转CSV

Convert PDF tables to CSV:

Write to

/tmp/request.json

json

{
  "url": "https://pdfco-test-files.s3.us-west-2.amazonaws.com/pdf-to-csv/sample.pdf",
  "inline": true
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/pdf/convert/to/csv" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

将PDF中的表格转换为CSV：

写入到

/tmp/request.json

：

json

{
  "url": "https://pdfco-test-files.s3.us-west-2.amazonaws.com/pdf-to-csv/sample.pdf",
  "inline": true
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/pdf/convert/to/csv" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

3. Merge PDFs

3. 合并PDF

Combine multiple PDFs into one:

Write to

/tmp/request.json

json

{
  "url": "https://pdfco-test-files.s3.us-west-2.amazonaws.com/pdf-merge/sample1.pdf,https://pdfco-test-files.s3.us-west-2.amazonaws.com/pdf-merge/sample2.pdf",
  "name": "merged.pdf"
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/pdf/merge" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

将多个PDF合并为一个文件：

写入到

/tmp/request.json

：

json

{
  "url": "https://pdfco-test-files.s3.us-west-2.amazonaws.com/pdf-merge/sample1.pdf,https://pdfco-test-files.s3.us-west-2.amazonaws.com/pdf-merge/sample2.pdf",
  "name": "merged.pdf"
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/pdf/merge" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

4. Split PDF

4. 拆分PDF

Split PDF by page ranges:

Write to

/tmp/request.json

json

{
  "url": "https://pdfco-test-files.s3.us-west-2.amazonaws.com/pdf-split/sample.pdf",
  "pages": "1-2,3-"
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/pdf/split" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

按页码范围拆分PDF：

写入到

/tmp/request.json

：

json

{
  "url": "https://pdfco-test-files.s3.us-west-2.amazonaws.com/pdf-split/sample.pdf",
  "pages": "1-2,3-"
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/pdf/split" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

5. Compress PDF

5. 压缩PDF

Reduce PDF file size:

Write to

/tmp/request.json

json

{
  "url": "https://pdfco-test-files.s3.us-west-2.amazonaws.com/pdf-optimize/sample.pdf",
  "name": "compressed.pdf"
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/pdf/optimize" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

减小PDF文件体积：

写入到

/tmp/request.json

：

json

{
  "url": "https://pdfco-test-files.s3.us-west-2.amazonaws.com/pdf-optimize/sample.pdf",
  "name": "compressed.pdf"
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/pdf/optimize" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

6. HTML to PDF

6. HTML转PDF

Convert HTML or URL to PDF:

Write to

/tmp/request.json

json

{
  "html": "<h1>Hello World</h1><p>This is a test.</p>",
  "name": "output.pdf"
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/pdf/convert/from/html" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

From URL:

Write to

/tmp/request.json

json

{
  "url": "https://example.com",
  "name": "webpage.pdf"
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/pdf/convert/from/url" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

将HTML或URL转换为PDF：

写入到

/tmp/request.json

：

json

{
  "html": "<h1>Hello World</h1><p>This is a test.</p>",
  "name": "output.pdf"
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/pdf/convert/from/html" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

从URL转换：

写入到

/tmp/request.json

：

json

{
  "url": "https://example.com",
  "name": "webpage.pdf"
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/pdf/convert/from/url" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

7. AI Invoice Parser

7. AI发票解析

Extract structured data from invoices:

Write to

/tmp/request.json

json

{
  "url": "https://pdfco-test-files.s3.us-west-2.amazonaws.com/ai-invoice-parser/sample-invoice.pdf",
  "inline": true
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/ai-invoice-parser" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

从发票中提取结构化数据：

写入到

/tmp/request.json

：

json

{
  "url": "https://pdfco-test-files.s3.us-west-2.amazonaws.com/ai-invoice-parser/sample-invoice.pdf",
  "inline": true
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/ai-invoice-parser" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

8. Upload Local File

8. 上传本地文件

Upload a local file first, then use the returned URL:

Step 1: Get presigned upload URL

bash

bash -c 'curl -s "https://api.pdf.co/v1/file/upload/get-presigned-url?name=myfile.pdf&contenttype=application/pdf" --header "x-api-key: ${PDFCO_API_KEY}"' | jq -r '.presignedUrl, .url'

Copy the presigned URL and file URL from the response.

Step 2: Upload file

Replace

<presigned-url>

with the URL from Step 1:

bash

curl -X PUT "<presigned-url>" --header "Content-Type: application/pdf" --data-binary @/path/to/your/file.pdf

Step 3: Use file URL in subsequent API calls

Replace

<file-url>

with the file URL from Step 1:

Write to

/tmp/request.json

json

{
  "url": "<file-url>",
  "inline": true
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/pdf/convert/to/text" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

先上传本地文件，再使用返回的URL进行处理：

步骤1：获取预签名上传URL

bash

bash -c 'curl -s "https://api.pdf.co/v1/file/upload/get-presigned-url?name=myfile.pdf&contenttype=application/pdf" --header "x-api-key: ${PDFCO_API_KEY}"' | jq -r '.presignedUrl, .url'

从响应中复制预签名URL和文件URL。

步骤2：上传文件

将

<presigned-url>

替换为步骤1获取的URL：

bash

curl -X PUT "<presigned-url>" --header "Content-Type: application/pdf" --data-binary @/path/to/your/file.pdf

步骤3：在后续API调用中使用文件URL

将

<file-url>

替换为步骤1获取的文件URL：

写入到

/tmp/request.json

：

json

{
  "url": "<file-url>",
  "inline": true
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/pdf/convert/to/text" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

9. Async Mode (Large Files)

9. 异步模式（适用于大文件）

For large files, use async mode to avoid timeouts:

Step 1: Start async job

Write to

/tmp/request.json

json

{
  "url": "https://example.com/large-file.pdf",
  "async": true
}

bash

bash -c 'curl -s --location --request POST "https://api.pdf.co/v1/pdf/convert/to/text" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json' | jq -r '.jobId'

Copy the job ID from the response.

Step 2: Check job status

Replace

<job-id>

with the job ID from Step 1:

Write to

/tmp/request.json

json

{
  "jobid": "<job-id>"
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/job/check" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

针对大文件使用异步模式避免超时：

步骤1：启动异步任务

写入到

/tmp/request.json

：

json

{
  "url": "https://example.com/large-file.pdf",
  "async": true
}

bash

bash -c 'curl -s --location --request POST "https://api.pdf.co/v1/pdf/convert/to/text" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json' | jq -r '.jobId'

从响应中复制任务ID。

步骤2：检查任务状态

将

<job-id>

替换为步骤1获取的任务ID：

写入到

/tmp/request.json

：

json

{
  "jobid": "<job-id>"
}

bash

bash -c 'curl --location --request POST "https://api.pdf.co/v1/job/check" --header "x-api-key: ${PDFCO_API_KEY}" --header "Content-Type: application/json" -d @/tmp/request.json'

Common Parameters

通用参数

Parameter	Type	Description
`url`	string	URL to source file (required)
`inline`	boolean	Return result in response body
`async`	boolean	Run as background job
`pages`	string	Page range, 1-indexed (e.g., "1-3", "1,3,5", "2-")
`name`	string	Output filename
`password`	string	PDF password if protected
`expiration`	integer	Output link expiration in minutes (default: 60)

参数	类型	说明
`url`	string	源文件的URL（必填）
`inline`	boolean	是否在响应体中直接返回结果
`async`	boolean	是否作为后台任务运行
`pages`	string	页码范围，从1开始计数（例如："1-3", "1,3,5", "2-"）
`name`	string	输出文件名
`password`	string	PDF的密码（如果文件受保护）
`expiration`	integer	输出链接的有效时长，单位为分钟（默认：60）

Response Format

响应格式

json

{
  "url": "https://pdf-temp-files.s3.amazonaws.com/.../result.pdf",
  "pageCount": 5,
  "error": false,
  "status": 200,
  "name": "result.pdf",
  "credits": 10,
  "remainingCredits": 9990
}

With

inline: true

, the response includes

body

field with extracted content.

json

{
  "url": "https://pdf-temp-files.s3.amazonaws.com/.../result.pdf",
  "pageCount": 5,
  "error": false,
  "status": 200,
  "name": "result.pdf",
  "credits": 10,
  "remainingCredits": 9990
}

当设置

inline: true

时，响应中会包含

body

字段，存放提取到的内容。

API Endpoints

API端点

Endpoint	Description
`/pdf/convert/to/text`	PDF to text (OCR supported)
`/pdf/convert/to/csv`	PDF to CSV
`/pdf/convert/to/json`	PDF to JSON
`/pdf/merge`	Merge multiple PDFs
`/pdf/split`	Split PDF by pages
`/pdf/optimize`	Compress PDF
`/pdf/convert/from/html`	HTML to PDF
`/pdf/convert/from/url`	URL to PDF
`/ai-invoice-parser`	AI-powered invoice parsing
`/document-parser`	Template-based document parsing
`/file/upload/get-presigned-url`	Get upload URL
`/job/check`	Check async job status

端点	说明
`/pdf/convert/to/text`	PDF转文本（支持OCR）
`/pdf/convert/to/csv`	PDF转CSV
`/pdf/convert/to/json`	PDF转JSON
`/pdf/merge`	合并多个PDF
`/pdf/split`	按页码拆分PDF
`/pdf/optimize`	压缩PDF
`/pdf/convert/from/html`	HTML转PDF
`/pdf/convert/from/url`	URL转PDF
`/ai-invoice-parser`	AI驱动的发票解析
`/document-parser`	基于模板的文档解析
`/file/upload/get-presigned-url`	获取上传URL
`/job/check`	检查异步任务状态

Guidelines

使用说明

File Sources: Use direct URLs or upload files first via presigned URL
Large Files: Use
```
async: true
```
for files over 40 pages or 10MB
OCR: Automatically enabled for scanned PDFs (set
```
lang
```
for non-English)
Rate Limits: Check your plan at https://pdf.co/pricing
Output Expiration: Download results within expiration time (default 60 min)
Credits: Each operation costs credits; check
```
remainingCredits
```
in response

文件来源：使用直接URL，或者先通过预签名URL上传文件
大文件处理：针对超过40页或10MB的文件，请设置
```
async: true
```
OCR：扫描版PDF会自动启用OCR（非英文内容请设置
```
lang
```
参数）
速率限制：可在https://pdf.co/pricing 查看你的套餐限制
输出有效期：请在有效期内下载结果（默认60分钟）
积分消耗：每次操作都会消耗积分，可在响应中查看
```
remainingCredits
```
剩余积分