5 Commits
4 changed files with 546 additions and 2 deletions
+4 -2
View File
@@ -2,7 +2,7 @@
A zero-dependency, single-file web dashboard for [OpenClaw](https://github.com/anthropics/openclaw). Manage agents, monitor token usage, and run commands — all from your browser, no `npm install` required.
[中文说明](#中文说明)
[中文说明](#中文说明) · [中文使用说明(含截图)](docs/USAGE_GUIDE.zh-CN.md) · [English Guide (with screenshots)](docs/USAGE_GUIDE.en.md)
---
@@ -48,7 +48,9 @@ Open [http://localhost:3333](http://localhost:3333) in your browser. For remote
**Model & Auth** — Configure models from your registered provider list. Follow built-in guides for provider authentication setup.
**Built-in CLI** — Run any `openclaw` command with real-time streaming output. Preset commands, favorites, and tab completion included.
**Built-in CLI** — Run any OpenClaw CLI command with real-time streaming output. Includes tab completion and saved favorites.
![](docs/redacted-screenshots/cli.jpg)
**Ops Panel** — Restart gateway, view logs, run health checks, manage backups (local + NAS via SFTP/rsync), and handle cron tasks.
+275
View File
@@ -0,0 +1,275 @@
# OCM Usage Guide (with screenshots)
> Audience: People who already have OpenClaw running locally (you can run `openclaw status` and read `openclaw gateway logs`).
>
> Goal: Use **OpenClaw Manager (OCM)** to make agent/sub-agent setup and day-to-day ops **visual, safer, and easier to audit**.
---
## 0) The pain: why OCM exists
Creating and maintaining OpenClaw agents is powerful, but the “default workflow” is often messy:
1) **Editing `openclaw.json` is risky**
- Deep/nested config, many fields.
- One accidental overwrite can break multiple agents (tokens, bindings, accountIds).
2) **Sub-agent topology is hard to see**
- With multiple bots + multiple groups, its not obvious which agent is bound to which Telegram peer.
3) **Model IDs are easy to misconfigure**
- A wrong `provider/model-id` can silently degrade performance, cause timeouts, or produce “typing…” with no reply.
4) **Usage and troubleshooting take time**
- Without tooling, youre digging through JSONL sessions and gateway logs to answer: “who used tokens?”, “what failed?”, “which provider?”
OCMs role: **a zero-dependency local UI** that helps you view and update OpenClaw configuration safely, keep changes auditable, and centralize common ops.
---
## 1) Why Telegram groups are the best way to build sub-agents
Think of it as:
> **One Telegram group = one agent boundary** (context + purpose + workspace).
### Benefits
1) **Clear isolation**
- Each agent speaks only inside its own group.
- Less cross-contamination between tasks.
2) **Controllable context**
- The group chat is the agents input stream.
- You can keep an agent “clean” by keeping the group focused.
3) **Natural delegation**
- A main agent can coordinate; sub-agents can specialize (tech / travel / finance / translation, etc.).
4) **Auditable**
- You can see exactly which message triggered which response.
### Safety / cost: keep groups private
Recommended rule:
- **Each agent group should contain only you + the bot** (and optionally your second account).
- Dont invite other people:
- cost: anyone can trigger token usage
- safety: higher risk of prompt-injection / accidental tool actions
---
## 2) Start OCM (macOS)
From the repo directory:
```bash
bash start.sh
```
Then open: <http://localhost:3333>
Restrict to localhost only:
```bash
bash start.sh --host 127.0.0.1
```
---
## 3) Dashboard: check health first
The Dashboard is your “sanity check”:
- CPU / RAM / Disk
- gateway running status
- agent counts
Screenshot:
![](redacted-screenshots/dashboard.jpg)
---
## 4) Create a main agent (with its own Telegram bot)
> Your `main` agent usually exists already. This section is for creating an additional **root agent** that owns its own Telegram bot/account.
### 4.1 BotFather prerequisites
1) Create a bot
- In Telegram, open **BotFather**
- Run `/newbot` → follow prompts → copy the bot token
2) Allow groups
- BotFather → your bot → Bot Settings
- **Allow Groups = ON**
3) Disable group privacy (critical)
- BotFather → your bot → Bot Settings → Group Privacy
- **Group Privacy = OFF**
If group privacy is ON, the bot cant see normal group messages. The symptom often looks like:
- slow replies
- “typing…” appears frequently
- sometimes no response at all
### 4.2 Add the agent in OCM
Go to **Agents**:
![](redacted-screenshots/agents.jpg)
Click `+ Add Agent` and fill:
- agent id / name
- workspace path (recommended: separate folder per agent)
- model (dropdown is sourced from `openclaw models list`)
- the BotFather token
OCM writes the agent + Telegram account/binding into `openclaw.json`.
---
## 5) Create a sub-agent (recommended workflow)
### 5.1 Why a new group per sub-agent
- clean context isolation
- easier debugging
- easy to “turn off” by muting a group
### 5.2 Step-by-step
In **Agents**`+ Add Sub-Agent`:
**Step 1 — Create a Telegram group**
- private group
- add only: you + the bot
**Step 2 — Get the Group ID (peer id)**
Use gateway logs:
```bash
openclaw gateway logs --follow
```
Send a message in the new group. In logs you should see something like:
- `-100xxxxxxxxxx` (Telegram group id)
**Step 3 — Fill the sub-agent form**
- choose the Parent Agent (which bot/account to share)
- paste the Group ID
- set workspace (recommended: its own folder)
- choose model
**Step 4 — Allowlist (optional but recommended)**
If you use `channels.telegram.allowFrom`, OCM can take “Your Telegram User ID” and append it automatically to the allowlist.
---
## 6) Verify bindings (Channels)
Go to **Channels**:
![](redacted-screenshots/channels.jpg)
Here you can quickly validate:
- which agent is bound to which Telegram peer
- whether `main` has the catch-all binding (if you use that pattern)
---
## 7) Model settings (Models)
Go to **Models**:
![](redacted-screenshots/models.jpg)
Key idea:
- OCM uses the real CLI model list (`openclaw models list`) for dropdowns.
- This reduces “invalid model-id” mistakes.
---
## 8) Credentials (Auth)
Go to **Auth**:
![](redacted-screenshots/auth.jpg)
Use this page to:
- see which provider profiles exist
- troubleshoot expired/missing tokens
---
## 9) Usage stats (Stats)
Go to **Stats**:
![](redacted-screenshots/stats.jpg)
You can break down usage by:
- model
- agent
- time window
OCM parses usage from OpenClaw session JSONL files.
---
## 10) Cron jobs (Cron)
Go to **Cron**:
![](redacted-screenshots/cron.jpg)
You can:
- see scheduled tasks (backup / update / health)
- run tasks manually
- enable/disable entries
---
## 11) Built-in CLI Terminal
OCM includes a built-in terminal panel to run OpenClaw CLI commands directly (for example: `openclaw status`, `openclaw gateway logs --follow`, `openclaw doctor`).
How to open: click **⌨️ Terminal** in the top navigation, or use the bottom CLI panel.
Highlights:
- **Tab completion** while typing commands
- **Common commands** dropdown to insert templates quickly
- **Favorites**: save frequently used commands and run them with one click
- **Streaming output** (stdout/stderr) for faster debugging
Screenshot:
![](redacted-screenshots/cli.jpg)
## 11) Fast troubleshooting
### “Typing…” but no reply in a Telegram group
Check (in order):
1) BotFather: **Group Privacy = OFF**
2) `openclaw status` (gateway healthy?)
3) provider auth / rate limit issues (Auth page + gateway logs)
### Model dropdown doesnt show a model you expect
- dropdown comes from `openclaw models list`
- ensure the model is registered/visible in your OpenClaw environment
---
## Notes about screenshots
All screenshots in this guide are **redacted** (no personal paths, no Telegram IDs).
They live under `docs/redacted-screenshots/`.
+267
View File
@@ -0,0 +1,267 @@
# OCM 功能使用说明(含截图)
> 面向:已经在本机安装并跑通 OpenClaw 的用户(至少会用 `openclaw status` / `openclaw gateway logs`
>
> 目标:用 **OpenClaw Manager (OCM)** 把“创建 Agent / Sub-Agent、绑定 Telegram 群、管理模型/认证、看用量、看 cron”这套流程做成可视化、可回溯、低风险。
---
## 0) 先说痛点:为什么要用 OCM
### 0.1 创建/维护 Agent 的典型痛点
1. **改 `openclaw.json` 容易改错**
- 字段多、层级深、容易手滑覆盖 token / accountId / bindings。
- 一次写错可能导致多个 Agent 全挂。
2. **Sub-Agent 的边界和“谁在用哪个 bot token”不直观**
- 多 Bot / 多 agent / 多群组绑定时,光靠肉眼看 JSON 很难确认拓扑结构是否正确。
3. **模型选择很容易“写了一个不存在的 model-id”**
- provider/model-id 拼错、或 provider 侧不支持,最后表现就是 Agent 变慢/报错/一直 typing。
4. **统计用量和排查问题耗时**
- 需要翻 session JSONL / gateway logs 才能知道谁在烧 token、哪里失败。
OCM 的定位:**把配置变更“可视化 + 可检查 + 可回滚(配合备份)”,并把常用运维入口集中起来。**
---
## 1) 为什么用 Telegram group 来建 Sub-Agent 最合适
> 这里讨论的是 OpenClaw 的一种非常实用的工作方式:**“一个 Telegram 群 = 一个 Agent 的工作空间/人格/上下文边界”**。
### 1.1 Telegram group 方案的优势
1. **边界天然清晰**
- 每个 Agent 只在自己的群里说话,和其他 Agent 隔离。
2. **上下文可控**
- 一个群的聊天记录就是该 Agent 的输入流;不会和别的任务混在一起。
3. **并发与分工自然**
- 主 Agent 负责总控/派工,Sub-Agent 负责专项:技术宅/旅行/理财/翻译……
4. **可审计**
- 群里发生了什么一眼可见;出问题也能定位是哪条消息触发。
### 1.2 为什么“群必须私密”(安全与成本)
- 强烈建议:**每个 Agent 群只留你自己 + 对应 bot(+ 必要时的另一个你自己的账号)**。
- 不要拉别人进来:
- 成本:别人一句话就触发 token 消耗
- 安全:可能诱导泄露信息/触发危险操作
---
## 2) 启动与进入 OCM
### 2.1 启动(macOS
在 OCM 目录:
```bash
bash start.sh
```
默认访问:<http://localhost:3333>
> 需要限制只能本机访问:`bash start.sh --host 127.0.0.1`
### 2.2 Dashboard(先看健康状态)
建议第一步先看 Dashboard,确认:
- 机器资源(CPU/RAM/DISK
- OpenClaw gateway 是否 Running
- Agents 数量
截图示例:
![](redacted-screenshots/dashboard.jpg)
---
## 3) 创建主 AgentMain Agent / 独立 Bot
> 你的 `main` 一般是默认存在的;这里重点讲“新增一个 **拥有独立 Telegram bot 的主 Agent**”。
### 3.1 BotFather 侧准备(必须)
1) 创建 bot
- 在 Telegram 找 **BotFather**
- 发送 `/newbot` → 按提示创建 → 拿到 bot token
2) 允许加群
- BotFather → 你的 bot → Bot Settings
- **Allow Groups = ON**
3) 关闭群隐私(非常关键)
- BotFather → 你的 bot → Bot Settings → Group Privacy
- **Group Privacy = OFF**
> 如果 Group Privacy 没关,bot 在群里看不到普通消息,会表现成“经常 typing 但没输出 / 没反应”。
### 3.2 在 OCM 添加主 Agent
进入 **Agents** 页面:
![](redacted-screenshots/agents.jpg)
点击 `+ Add Agent`,按表单填写:
- Agent id / name
- workspace 路径(建议独立目录)
- 选择模型(下拉来自 `openclaw models list`,更不容易选错)
- 填入刚拿到的 bot token
保存后,OCM 会把这颗主 Agent 写进 `openclaw.json`,并把 bot token 绑定到对应 `accountId`
---
## 4) 创建 Sub-Agent(最常用)
### 4.1 为什么 Sub-Agent 要用“新群”
- 让每个 Sub-Agent 有自己独立的对话上下文、任务边界
- 出问题直接在该群排查,不污染主 Agent 的对话
### 4.2 Sub-Agent 创建流程(逐步)
进入 **Agents** 页面 → 点击 `+ Add Sub-Agent`,按引导做:
#### Step 1:建 Telegram 群
- 新建群(私密)
- 群里只加:你自己 + 对应的 bot
#### Step 2:拿到群 IDPeer ID / Group ID
OpenClaw 通常会在 gateway 日志里打印群的 peer id。
常见拿法:
```bash
openclaw gateway logs --follow
```
然后在 Telegram 群里随便发一句话,观察日志。
> 你会看到类似 `-100xxxxxxxxxx` 这样的 group id。
#### Step 3:回到 OCM 填表
- Parent Agent(选择共享哪个 bot/account
- 填 Group ID
- workspace(建议单独目录)
- 选择模型
#### Step 4:写 allowlist(推荐)
如果你启用了 Telegram 的 allowlist(例如 `channels.telegram.allowFrom`),
OCM 的表单里可以填 “Your Telegram User ID”,让它自动把你的 user id 加进去,避免 DM/群消息被拒。
---
## 5) 绑定检查(Channels 页)
进入 **Channels**
![](redacted-screenshots/channels.jpg)
这里能一眼看到:
- 哪个 agent 绑定了哪个 Telegram peer(群)
- main 的“兜底绑定”(any)是否存在
> 如果某个 Sub-Agent 绑定错群/丢 binding,最直观的就是在这里对不上。
---
## 6) 模型管理(Models 页)
进入 **模型**
![](redacted-screenshots/models.jpg)
关键点:
- 主模型(primary)和 fallback 链
- 下拉列表来自 `openclaw models list`(能减少“选了不存在模型”的问题)
---
## 7) 认证管理(Auth 页)
进入 **认证**
![](redacted-screenshots/auth.jpg)
这里主要用来:
- 看当前有哪些 provider profile
- 排查某个 provider 是否过期/缺 token
---
## 8) 用量统计(Stats 页)
进入 **Stats**
![](redacted-screenshots/stats.jpg)
你可以按:
- 按模型聚合
- 按 Agent 聚合
- 按天趋势
> OCM 是从 OpenClaw 的 session JSONL 里解析真实 usage(比“看日志猜”靠谱)。
---
## 9) CLI 终端(内置)
OCM 内置一个 CLI 面板,用来直接运行 OpenClaw 命令(例如:`openclaw status``openclaw gateway logs --follow``openclaw doctor`)。
入口:点击顶部导航栏的 **⌨️ 终端**,或使用页面底部的 CLI 区域。
功能要点:
- **Tab 补全**:输入命令时按 Tab 进行补全/提示(适合 `openclaw ...` 子命令)
- **常用命令**:可从下拉选择,一键填充常用命令模板
- **收藏(Favorites)**:把常用命令保存为收藏,后续一键执行
- **实时输出**:执行后会流式显示 stdout/stderr,方便排错
截图示例:
![](redacted-screenshots/cli.jpg)
## 9) Cron(定时任务)
进入 **Cron**
![](redacted-screenshots/cron.jpg)
用途:
- 查看有哪些健康检查/更新/备份任务
- 手动运行
- 观察启用/禁用状态
---
## 10) 常见故障排查(快速)
### 10.1 Telegram 群里“一直 typing 但不回”
优先检查:
1) BotFather 的 **Group Privacy 是否 OFF**
2) OpenClaw gateway 是否健康(`openclaw status`
3) provider 是否限流/超时(Auth 页 + gateway logs
### 10.2 模型下拉没有你想要的模型
- 现在下拉是 `openclaw models list` 的真实输出。
- 先确保你在 OpenClaw 里已经注册/可见该模型。
---
## 11) 推荐的“最佳实践”
- 每个 Agent 都用独立 workspace(隔离文件/记忆)
- 每个 Sub-Agent 都用独立 Telegram 群(隔离上下文)
- 群里不要加第三方(成本 + 安全)
- 任何“改配置的大动作”尽量先备份(OCM/脚本/手工都行)
---
## 附:截图说明
本文截图来自 `docs/redacted-screenshots/`,已做脱敏(路径用户名、Telegram 绑定 ID 等均已模糊处理)。
Binary file not shown.

After

Width:  |  Height:  |  Size: 240 KiB