ralphing

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Ralphing

运行Ralph自主编码循环

Ralph is an autonomous AI coding loop that ships features while you sleep. Each iteration runs in a fresh context window, while memory persists through git history and text files.

Ralph是一款自主AI编码循环工具，能在你休息时完成功能开发并交付。每次迭代都会在全新的上下文窗口中运行，而记忆则通过git历史记录和文本文件得以保留。

When to Use

使用场景

When the user wants to implement a multi-story feature autonomously, or mentions "ralph", "ralphing", or "autonomous loop".

当用户希望自主实现多需求故事的功能，或者提到“ralph”、“ralphing”或“自主循环”时使用。

Workflow

工作流程

Step 1: Get the PRD

步骤1：获取PRD文件

The user provides a PRD file, or you help them create one. Ask:

Do you have a PRD already, or should we create one together?
What's the repo path?
What's the test command? (e.g.,
```
bundle exec rake
```
,
```
npm test
```
)
How many iterations max? (default: 25)

用户提供PRD文件，或者你协助他们创建。请询问：

你已有PRD文件，还是需要我们一起创建？
仓库路径是什么？
测试命令是什么？（例如：
```
bundle exec rake
```
、
```
npm test
```
）
最大迭代次数是多少？（默认值：25）

PRD Format

PRD格式

markdown

undefined

markdown

undefined

PRD

Branch:

<branch-name>

Branch:

<branch-name>

Stories

US-001: <Story title>

<Acceptance criterion>
<Acceptance criterion>

<Acceptance criterion>
<Acceptance criterion>

US-002: <Story title>

<Acceptance criterion>


Guidelines for stories:
- **Small**: Must fit in one context window
- **Explicit criteria**: Avoid vague ("Users can log in"), prefer specific checks
- Story order = priority (first = highest)

Note: Ralph is typically run in a git worktree. If already on the correct branch (or detached HEAD), the branch checkout is skipped.

<Acceptance criterion>


需求故事编写准则：
- **粒度要小**：必须能容纳在一个上下文窗口中
- **验收标准明确**：避免模糊表述（如“用户可以登录”），优先使用可具体验证的检查项
- 故事顺序=优先级（第一个故事优先级最高）

注意：Ralph通常在git worktree中运行。如果已处于正确分支（或分离HEAD状态），则会跳过分支检出步骤。

Step 2: Start the Loop

步骤2：启动循环

bash

~/.claude/skills/ralphing/assets/ralph.sh start <repo_path> <prd.md> [max_iterations] [test_cmd]

This returns a session directory (e.g.,

/tmp/ralph/ABC123-...

) and runs the loop in the background.

bash

~/.claude/skills/ralphing/assets/ralph.sh start <repo_path> <prd.md> [max_iterations] [test_cmd]

这会生成一个会话目录（例如：

/tmp/ralph/ABC123-...

）并在后台运行循环。

Step 3: Monitor Progress

步骤3：监控进度

Poll the session status periodically:

bash

~/.claude/skills/ralphing/assets/ralph.sh status <ralph_dir>

This outputs:

```
status
```
:
```
running
```
,
```
complete
```
, or
```
failed
```
```
iteration
```
: current/max iterations
Full contents of
```
progress.txt
```

Keep polling until status is

complete

failed

定期轮询会话状态：

bash

~/.claude/skills/ralphing/assets/ralph.sh status <ralph_dir>

该命令会输出：

```
status
```
：
```
running
```
（运行中）、
```
complete
```
（已完成）或
```
failed
```
（失败）
```
iteration
```
：当前/最大迭代次数
```
progress.txt
```
的完整内容

持续轮询直到状态变为

complete

或

failed

。

How the Loop Works

循环工作原理

```
start
```
copies PRD, generates prompt, checks out branch (if needed), and spawns background process
Background loop: agent implements story -> commits -> runs tests
If tests fail, reverts commit and retries (progress.txt preserved)
Exits when agent signals
```
<promise>COMPLETE</promise>
```
or max iterations reached
Parent agent polls
```
status
```
to observe progress without blocking

```
start
```
命令会复制PRD、生成提示词、检出分支（如有需要）并启动后台进程
后台循环流程：Agent实现需求故事 -> 提交代码 -> 运行测试
如果测试失败，会回滚提交并重试（
```
progress.txt
```
内容会保留）
当Agent发出
```
<promise>COMPLETE</promise>
```
信号或达到最大迭代次数时，循环退出
父Agent通过轮询
```
status
```
来观察进度，不会造成阻塞

ralphing

Original

Translation

Ralphing

运行Ralph自主编码循环

When to Use

使用场景

Workflow

工作流程

Step 1: Get the PRD

步骤1：获取PRD文件

PRD Format

PRD格式

PRD

PRD

Stories

Stories

US-001: <Story title>

US-001: <Story title>

US-002: <Story title>

US-002: <Story title>

Step 2: Start the Loop

步骤2：启动循环

Step 3: Monitor Progress

步骤3：监控进度

How the Loop Works

循环工作原理

Resources

相关资源