# openclaw-auto-continue

可重用的 OpenClaw skill，提供 `/auto-continue`（別名 `/ac`）的 session-level auto-followup 使用方式。

Reusable OpenClaw skill for `/auto-continue` (`/ac`) session-level auto-followup.

## 繁體中文說明

### 這個 repo 內有什麼
- `openclaw-auto-continue/SKILL.md` — **給 Agent 看的** skill 指引
- `openclaw-auto-continue/references/installation-guide.md` — **給要落地 `/ac` 指令的人 / agent 看** 的安裝指南
- `openclaw-auto-continue/references/runtime-integration.md` — `/ac` runtime integration 說明
- `openclaw-auto-continue.skill` — 已打包好的 skill 檔

### 其他電腦怎麼用

#### 方式 1：直接 clone repo
```bash
git clone https://gitea.cowbay.org/openclaw/openclaw-auto-continue.git
cd openclaw-auto-continue
```

把 skill 資料夾放到目標機器的 OpenClaw skills 目錄：

```bash
mkdir -p ~/.openclaw/workspace/skills
cp -R openclaw-auto-continue ~/.openclaw/workspace/skills/
```

#### 方式 2：直接用 `.skill` 檔
如果目標 OpenClaw 環境已有 skill 匯入 / 安裝流程，就直接使用：

- `openclaw-auto-continue.skill`

### 在目標 OpenClaw 聊天 / session 裡怎麼開
安裝 skill **不代表每個 chat 都會自動啟用**。
你還是要在想啟用的那個 chat / session 裡手動開：

```text
/ac on
```

查狀態：

```text
/ac status
```

關閉：

```text
/ac off
```

### `/ac on` 之後會發生什麼
開啟後，目前 session 會使用 followup 型續跑設定：

- `queueMode = "followup"`
- `queueDebounceMs = 0`
- `queueCap = 1`
- `queueDrop = "summarize"`

實際效果：
- agent 比較可能直接接著做下一個低風險步驟
- 比較不容易每做一步就停下來
- 適合在同一個 chat 中做多步任務

### 重要邊界
這是 **session-scoped**，不是全域開關。  
如果你想在另一個 chat 生效，就要在那個 chat 再開一次。

它也**不會**繞過：
- approvals
- safety checks
- destructive action confirmation
- user decision points

### 如果其他電腦出現 `/ac` unknown
這通常代表目標機器的 **live OpenClaw runtime 還沒有真正實作這個 command**。

所以這個 repo 不只要教怎麼用，還要教怎麼把 `/ac` 接進 runtime。
請看：

- `openclaw-auto-continue/references/runtime-integration.md`

這份文件會說明：
- runtime handler 應該做什麼
- `/ac status`、`/ac on`、`/ac off` 應該如何改變 session
- 如何驗證另一台真實機器
- 如果 skill 存在但 command 不存在，應該怎麼排查

### 給 Agent 看
如果 OpenClaw agent 要判斷怎麼使用這個 repo：

1. 先讀 `openclaw-auto-continue/SKILL.md`
2. 如果目標機器不認得 `/ac`，再讀 `openclaw-auto-continue/references/runtime-integration.md`
3. 在對應 chat / session 中使用 `/ac on`、`/ac off`、`/ac status`

一句話：
- `README.md` 是給人 / operator 看的
- `openclaw-auto-continue/SKILL.md` 是給 agent 看的
- `references/installation-guide.md` 是給要從頭做到尾落地 `/ac` 的人 / agent 看的
- `references/runtime-integration.md` 是給要理解 runtime 接法的人 / agent 看的

---

## English

### What this repo contains
- `openclaw-auto-continue/SKILL.md` — the **agent-facing** skill instructions
- `openclaw-auto-continue/references/installation-guide.md` — end-to-end guide for making `/ac` real on another machine
- `openclaw-auto-continue/references/runtime-integration.md` — runtime integration notes for `/ac`
- `openclaw-auto-continue.skill` — packaged skill artifact

### For humans: how to use on another computer

#### Option 1: clone the repo
```bash
git clone https://gitea.cowbay.org/openclaw/openclaw-auto-continue.git
cd openclaw-auto-continue
```

Then copy the skill folder into the target machine's OpenClaw skills directory:

```bash
mkdir -p ~/.openclaw/workspace/skills
cp -R openclaw-auto-continue ~/.openclaw/workspace/skills/
```

#### Option 2: use the packaged `.skill` file
If your target OpenClaw environment already has a skill import/install flow, use:

- `openclaw-auto-continue.skill`

### On the target OpenClaw chat/session
Installing the skill does **not** automatically enable auto-continue for every chat.
You still need to turn it on in the specific chat/session where you want it active:

```text
/ac on
```

Check status:

```text
/ac status
```

Turn it off:

```text
/ac off
```

### What changes after `/ac on`
When enabled, the current session uses followup-style continuation behavior:

- `queueMode = "followup"`
- `queueDebounceMs = 0`
- `queueCap = 1`
- `queueDrop = "summarize"`

Practical effect:
- the agent is more likely to continue into the next low-risk step
- the agent is less likely to stop after every intermediate update
- this is good for multi-step execution in the same chat

### Important boundary
This is **session-scoped**, not global.
If you want it active in a different chat, enable it in that chat too.

It also does **not** bypass:
- approvals
- safety checks
- destructive action confirmation
- user decision points

### If another machine says `/ac` is unknown
That usually means the target machine's **live OpenClaw runtime does not actually implement the command yet**.

So this repo should teach not only usage, but also how to integrate `/ac` into the runtime.
See:

- `openclaw-auto-continue/references/runtime-integration.md`

That file explains:
- what the runtime handler must do
- what `/ac status`, `/ac on`, `/ac off` should change
- how to verify a real target machine
- how to troubleshoot a machine where the skill exists but the command does not

### For agents
If an OpenClaw agent is deciding how to use this repo:

1. Read `openclaw-auto-continue/SKILL.md`
2. If the target machine does not recognize `/ac`, read `openclaw-auto-continue/references/runtime-integration.md`
3. Use `/ac on`, `/ac off`, and `/ac status` in the relevant chat/session

In short:
- `README.md` is the human/operator guide
- `openclaw-auto-continue/SKILL.md` is the agent-facing guide
- `references/installation-guide.md` is for end-to-end `/ac` installation work
- `references/runtime-integration.md` is for runtime integration details