openai-cli

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

OpenAI CLI

Skill by ara.so — Devtools Skills collection.

The official CLI for the OpenAI REST API. Provides a command-line interface to interact with OpenAI's API endpoints including chat completions, embeddings, fine-tuning, and admin operations.

由ara.so提供的技能——Devtools技能合集。

OpenAI REST API的官方CLI。提供命令行界面，用于与OpenAI的API端点交互，包括聊天补全、嵌入、微调以及管理操作。

Installation

安装

Homebrew (macOS/Linux)

Homebrew（macOS/Linux）

bash

brew install openai/tools/openai

bash

brew install openai/tools/openai

Go Install

Go 安装

bash

go install 'github.com/openai/openai-cli/cmd/openai@latest'

Ensure Go bin is in your PATH:

bash

export PATH="$PATH:$(go env GOPATH)/bin"

bash

go install 'github.com/openai/openai-cli/cmd/openai@latest'

确保Go的bin目录在你的PATH中：

bash

export PATH="$PATH:$(go env GOPATH)/bin"

Local Development

本地开发

Clone the repository and run locally:

bash

git clone https://github.com/openai/openai-cli.git
cd openai-cli
./scripts/run args...

克隆仓库并本地运行：

bash

git clone https://github.com/openai/openai-cli.git
cd openai-cli
./scripts/run args...

Configuration

配置

Set up your API credentials using environment variables:

bash

export OPENAI_API_KEY="sk-..."
export OPENAI_ADMIN_KEY="sk-admin-..."  # For admin endpoints
export OPENAI_ORG_ID="org-..."          # Optional
export OPENAI_PROJECT_ID="proj_..."     # Optional

Or use flags on each command:

bash

openai --api-key sk-... responses create --input "Hello"

使用环境变量设置你的API凭据：

bash

export OPENAI_API_KEY="sk-..."
export OPENAI_ADMIN_KEY="sk-admin-..."  # 用于管理端点
export OPENAI_ORG_ID="org-..."          # 可选
export OPENAI_PROJECT_ID="proj_..."     # 可选

或者在每个命令中使用参数：

bash

openai --api-key sk-... responses create --input "Hello"

Environment Variables

环境变量

Variable	Purpose	Required
`OPENAI_API_KEY`	Standard API authentication	For standard endpoints
`OPENAI_ADMIN_KEY`	Admin API authentication	For admin endpoints
`OPENAI_ORG_ID`	Organization identifier	No
`OPENAI_PROJECT_ID`	Project identifier	No
`OPENAI_WEBHOOK_SECRET`	Webhook verification	No

变量	用途	是否必填
`OPENAI_API_KEY`	标准API认证	用于标准端点
`OPENAI_ADMIN_KEY`	管理API认证	用于管理端点
`OPENAI_ORG_ID`	组织标识符	否
`OPENAI_PROJECT_ID`	项目标识符	否
`OPENAI_WEBHOOK_SECRET`	Webhook验证	否

Command Structure

命令结构

bash

openai [resource] <command> [flags...]

bash

openai [resource] <command> [flags...]

Key Commands

核心命令

Chat Completions

聊天补全

Create a chat completion:

bash

openai responses create \
  --input "Explain quantum computing in simple terms" \
  --model gpt-4

Stream responses:

bash

openai responses create \
  --input "Write a poem about coding" \
  --model gpt-4 \
  --stream

With system message and temperature:

bash

openai responses create \
  --input "What is recursion?" \
  --model gpt-4 \
  --system "You are a computer science teacher" \
  --temperature 0.7

创建聊天补全：

bash

openai responses create \
  --input "用简单的语言解释量子计算" \
  --model gpt-4

流式响应：

bash

openai responses create \
  --input "写一首关于编程的诗" \
  --model gpt-4 \
  --stream

携带系统消息和温度参数：

bash

openai responses create \
  --input "什么是递归？" \
  --model gpt-4 \
  --system "你是一名计算机科学老师" \
  --temperature 0.7

File Operations

文件操作

Upload a file:

bash

openai files create \
  --file @training_data.jsonl \
  --purpose fine-tune

List files:

bash

openai files list

Retrieve file content:

bash

openai files retrieve file-abc123

Delete a file:

bash

openai files delete file-abc123

上传文件：

bash

openai files create \
  --file @training_data.jsonl \
  --purpose fine-tune

列出文件：

bash

openai files list

获取文件内容：

bash

openai files retrieve file-abc123

删除文件：

bash

openai files delete file-abc123

Embeddings

嵌入

Generate embeddings:

bash

openai embeddings create \
  --input "The quick brown fox" \
  --model text-embedding-3-small

Multiple inputs:

bash

openai embeddings create \
  --input "First text" \
  --input "Second text" \
  --model text-embedding-3-large

生成嵌入：

bash

openai embeddings create \
  --input "The quick brown fox" \
  --model text-embedding-3-small

多输入：

bash

openai embeddings create \
  --input "First text" \
  --input "Second text" \
  --model text-embedding-3-large

Fine-tuning

微调

Create a fine-tuning job:

bash

openai fine-tuning jobs create \
  --training-file file-abc123 \
  --model gpt-3.5-turbo

List fine-tuning jobs:

bash

openai fine-tuning jobs list

Check job status:

bash

openai fine-tuning jobs retrieve ftjob-abc123

Cancel a job:

bash

openai fine-tuning jobs cancel ftjob-abc123

创建微调任务：

bash

openai fine-tuning jobs create \
  --training-file file-abc123 \
  --model gpt-3.5-turbo

列出微调任务：

bash

openai fine-tuning jobs list

查看任务状态：

bash

openai fine-tuning jobs retrieve ftjob-abc123

取消任务：

bash

openai fine-tuning jobs cancel ftjob-abc123

Admin Operations

管理操作

Usage statistics:

bash

openai admin:organization:usage completions \
  --start-time 1735689600 \
  --end-time 1735776000 \
  --bucket-width 1d

使用统计：

bash

openai admin:organization:usage completions \
  --start-time 1735689600 \
  --end-time 1735776000 \
  --bucket-width 1d

File Handling

文件处理

Passing Files as Arguments

传递文件作为参数

Use

prefix for file paths:

bash

openai responses create --arg @input.txt

Within JSON/YAML:

bash

openai <command> --arg '{image: "@image.jpg"}'

bash

openai <command> <<YAML
arg:
  image: "@abe.jpg"
YAML

使用

前缀表示文件路径：

bash

openai responses create --arg @input.txt

在JSON/YAML中：

bash

openai <command> --arg '{image: "@image.jpg"}'

bash

openai <command> <<YAML
arg:
  image: "@abe.jpg"
YAML

Explicit Encoding

显式编码

Force string encoding:

bash

openai <command> --arg @file://myfile.txt

Force base64 encoding:

bash

openai <command> --arg @data://image.png

Absolute paths:

bash

openai <command> --arg @file:///tmp/file.txt

强制字符串编码：

bash

openai <command> --arg @file://myfile.txt

强制base64编码：

bash

openai <command> --arg @data://image.png

绝对路径：

bash

openai <command> --arg @file:///tmp/file.txt

Escaping @ Signs

转义@符号

To pass a literal

character:

bash

openai <command> --username '\@username'

要传递字面量

字符：

bash

openai <command> --username '\@username'

Output Formatting

输出格式化

Format Options

格式选项

Control output format with

--format

bash

undefined

使用

--format

控制输出格式：

bash

undefined

Pretty-printed (default for TTY)

美观打印（TTY默认）

openai responses create --input "Hello" --format pretty

Raw JSON

原始JSON

openai responses create --input "Hello" --format json

JSONL (one JSON object per line)

JSONL（每行一个JSON对象）

openai responses create --input "Hello" --format jsonl

YAML

openai responses create --input "Hello" --format yaml

Raw response body

原始响应体

openai responses create --input "Hello" --format raw

undefined

openai responses create --input "Hello" --format raw

undefined

Data Transformation

数据转换

Use GJSON syntax to extract specific fields:

bash

undefined

使用GJSON语法提取特定字段：

bash

undefined

Get just the message content

仅获取消息内容

openai responses create
--input "Hello"
--transform "choices.0.message.content"

Get multiple fields

获取多个字段

openai responses create
--input "Hello"
--transform "{id: id, content: choices.0.message.content}"


Transform errors:

```bash
openai responses create \
  --input "Hello" \
  --format-error json \
  --transform-error "error.message"

openai responses create
--input "Hello"
--transform "{id: id, content: choices.0.message.content}"


转换错误信息：

```bash
openai responses create \
  --input "Hello" \
  --format-error json \
  --transform-error "error.message"

Global Flags

全局参数

bash

--api-key           # API key (or use OPENAI_API_KEY)
--admin-api-key     # Admin key (or use OPENAI_ADMIN_KEY)
--organization      # Org ID (or use OPENAI_ORG_ID)
--project           # Project ID (or use OPENAI_PROJECT_ID)
--base-url          # Custom API endpoint
--debug             # Enable debug logging with HTTP details
--version, -v       # Show CLI version
--help              # Show help for command
--format            # Output format (auto, explore, json, jsonl, pretty, raw, yaml)
--format-error      # Error output format
--transform         # GJSON transform for data output
--transform-error   # GJSON transform for error output

bash

--api-key           # API密钥（或使用OPENAI_API_KEY）
--admin-api-key     # 管理密钥（或使用OPENAI_ADMIN_KEY）
--organization      # 组织ID（或使用OPENAI_ORG_ID）
--project           # 项目ID（或使用OPENAI_PROJECT_ID）
--base-url          # 自定义API端点
--debug             # 启用包含HTTP详情的调试日志
--version, -v       # 显示CLI版本
--help              # 显示命令帮助
--format            # 输出格式（auto, explore, json, jsonl, pretty, raw, yaml）
--format-error      # 错误输出格式
--transform         # 数据输出的GJSON转换
--transform-error   # 错误输出的GJSON转换

Common Patterns

常见使用模式

Interactive Chat Session

交互式聊天会话

bash

while true; do
  read -p "You: " input
  openai responses create \
    --input "$input" \
    --model gpt-4 \
    --format pretty
done

bash

while true; do
  read -p "你: " input
  openai responses create \
    --input "$input" \
    --model gpt-4 \
    --format pretty
done

Batch Processing

批量处理

Process multiple inputs from a file:

bash

while IFS= read -r line; do
  openai responses create \
    --input "$line" \
    --model gpt-3.5-turbo \
    --format json >> results.jsonl
done < inputs.txt

处理文件中的多个输入：

bash

while IFS= read -r line; do
  openai responses create \
    --input "$line" \
    --model gpt-3.5-turbo \
    --format json >> results.jsonl
done < inputs.txt

JSON Input

JSON输入

Pass structured data:

bash

openai responses create \
  --input "Translate to French" \
  --model gpt-4 \
  --max-tokens 100 \
  --temperature 0.3

传递结构化数据：

bash

openai responses create \
  --input "翻译成法语" \
  --model gpt-4 \
  --max-tokens 100 \
  --temperature 0.3

Piping Data

管道传递数据

bash

echo "Summarize this text" | openai responses create \
  --input "$(cat)" \
  --model gpt-4

Or:

bash

cat document.txt | openai responses create \
  --input "Summarize: $(cat -)" \
  --model gpt-4

bash

echo "总结这段文字" | openai responses create \
  --input "$(cat)" \
  --model gpt-4

或者：

bash

cat document.txt | openai responses create \
  --input "总结：$(cat -)" \
  --model gpt-4

Image Analysis

图像分析

bash

openai responses create \
  --input '{text: "What is in this image?", image: "@photo.jpg"}' \
  --model gpt-4-vision-preview

bash

openai responses create \
  --input '{text: "这张图片里有什么？", image: "@photo.jpg"}' \
  --model gpt-4-vision-preview

Custom Base URL

自定义基础URL

For Azure or custom endpoints:

bash

openai --base-url https://custom.openai.azure.com/v1 \
  responses create \
  --input "Hello" \
  --model gpt-4

针对Azure或自定义端点：

bash

openai --base-url https://custom.openai.azure.com/v1 \
  responses create \
  --input "Hello" \
  --model gpt-4

Debugging

调试

Enable debug mode to see full HTTP requests/responses:

bash

openai --debug responses create --input "Test"

Warning: Debug logs include request/response bodies which may contain sensitive data.

启用调试模式查看完整的HTTP请求/响应：

bash

openai --debug responses create --input "Test"

警告：调试日志包含请求/响应体，可能包含敏感数据。

Troubleshooting

故障排除

Command Not Found

命令未找到

Ensure Go bin is in your PATH:

bash

echo $PATH | grep "$(go env GOPATH)/bin"

确保Go的bin目录在你的PATH中：

bash

echo $PATH | grep "$(go env GOPATH)/bin"

If not found:

如果未找到：

export PATH="$PATH:$(go env GOPATH)/bin"

undefined

export PATH="$PATH:$(go env GOPATH)/bin"

undefined

Authentication Errors

认证错误

Verify API key is set:

bash

echo $OPENAI_API_KEY

验证API密钥已设置：

bash

echo $OPENAI_API_KEY

Should output: sk-...

应输出：sk-...


Check key validity:

```bash
openai responses create --input "test" --model gpt-3.5-turbo


检查密钥有效性：

```bash
openai responses create --input "test" --model gpt-3.5-turbo

File Not Found Errors

文件未找到错误

Use absolute paths or verify relative paths:

bash

undefined

使用绝对路径或验证相对路径：

bash

undefined

Absolute path

绝对路径

openai files create --file @/full/path/to/file.jsonl --purpose fine-tune

Verify file exists

验证文件存在

ls -la @file.jsonl

undefined

ls -la @file.jsonl

undefined

Rate Limiting

速率限制

Add delays between requests:

bash

for i in {1..10}; do
  openai responses create --input "Request $i"
  sleep 1
done

在请求之间添加延迟：

bash

for i in {1..10}; do
  openai responses create --input "Request $i"
  sleep 1
done

Model Not Available

模型不可用

List available models (if endpoint exists):

bash

openai models list

Check model name spelling and access permissions.

列出可用模型（如果端点存在）：

bash

openai models list

检查模型名称拼写和访问权限。

Output Not Formatted

输出未格式化

Force format explicitly:

bash

openai responses create --input "Hello" --format json

Check if output is being piped (auto-detection may choose raw format):

bash

openai responses create --input "Hello" --format pretty | less

显式强制格式：

bash

openai responses create --input "Hello" --format json

检查输出是否被管道传输（自动检测可能选择原始格式）：

bash

openai responses create --input "Hello" --format pretty | less

Linking Custom SDK Versions

链接自定义SDK版本

For development, link against different Go SDK versions:

bash

undefined

开发时，链接不同的Go SDK版本：

bash

undefined

Link to specific version

链接到特定版本

./scripts/link github.com/openai/openai-go@v1.2.3

Link to local SDK copy

链接到本地SDK副本

./scripts/link ../openai-go

Default (../openai-go)

默认（../openai-go）

./scripts/link

undefined

./scripts/link

undefined

Version Information

版本信息

Check CLI version:

bash

openai --version

检查CLI版本：

bash

openai --version

or

或

openai -v


Get help:

```bash
openai --help
openai responses --help
openai responses create --help

openai -v


获取帮助：

```bash
openai --help
openai responses --help
openai responses create --help