release: v1.0.0-rc1 — feature freeze for stable release

* fix: orphan agents now grouped under main in agent tree

Agents without parentAgentId, own bot, or binding-based inference
(e.g. dynamically created sub-agents like codex-worker) were
incorrectly displayed as top-level roots. They now default to
main's subtree.

Bump version to 0.9.4.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* cleanup: drop archived workflow skill defaults

* release: v1.0.0-rc1 — feature freeze for stable release

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: dtzp555 <dtzp555@gmail.com>

DEVLOG.md

@@ -1,10 +1,35 @@
 # OpenClaw Manager — 开发日志
 
-> 最后更新：2026-03-01
+> 最后更新：2026-03-17
-> 当前版本：v0.9.0
+> 当前版本：v0.9.4
 
 ---
 
+## v0.9.4 更新日志（2026-03-17）
+
+### New Features
+
+- **Skills & Tools 自动配置**：创建 Agent / Sub-Agent 时自动加载默认 Skills 和 Tool Groups
+  - 默认 Skills：`memory-continuity`, `agent-workflow`, `execution-agent-dispatch`, `session-logs`
+  - 默认 Tool Groups：`group:fs`, `group:runtime`, `group:memory`, `sessions_spawn`, `subagents`
+  - 创建表单新增可折叠的 Skills & Tools 选择面板，支持 checkbox 多选
+  - 提供「仅默认」「全选」「清空」快捷按钮
+  - 不展开 / 不修改则自动使用默认预设，开箱即用
+- 支持的可选 Skills：browser-use, github, gh-issues, coding-agent, execution-agent-planner, discord, weather, summarize, healthcheck
+- 四个创建端点（Telegram Agent/Sub、Discord Agent/Sub）均支持 `skills` + `toolGroups` 字段
+
+### Backend
+
+- 新增 `applySkillsTools()` 服务端辅助函数，统一处理 skills/tools 的默认值回退逻辑
+- `POST /api/agents/bot`、`POST /api/agents/discord`、`POST /api/agents/discord-sub`、`POST /api/agents` 均已适配
+
+### UI
+
+- Skills & Tools Picker：深色风格 checkbox 网格，嵌入创建表单底部
+- 新增专用 CSS 样式（`.skills-picker`、`.sp-grid`、`.sp-item`、`.sp-actions`）
+- 新增 i18n 字符串支持中英文标签
+
+---
 
 ## v0.9.0 更新日志（2026-03-01）
 

README.md

@@ -1,50 +1,74 @@
 # OpenClaw Manager (OCM)
 
-A zero-dependency local control panel for [OpenClaw](https://github.com/anthropics/openclaw). Manage agents, bindings, logs, commands, and rollback from your browser — no `npm install` required.
+A local control panel for [OpenClaw](https://github.com/anthropics/openclaw) that makes setup checks, agent management, health inspection, and recovery much easier — without living in JSON and terminal tabs all day.
 
 [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)
 
 ---
 
-## Quick Start
+## Why OCM exists
 
-```bash
+Running OpenClaw directly is powerful, but the day-to-day workflow can get messy fast:
-# Clone
-git clone https://github.com/dtzp555-max/ocm.git
-cd ocm
 
-# Run (no npm install needed)
+- `openclaw.json` is easy to break by hand
-bash start.sh        # macOS / Linux
+- agent / bot / channel bindings are hard to visualize
-start.bat            # Windows
+- health checks, logs, restarts, and rollback are scattered across terminal commands
+- when something goes wrong, recovery is slower than it should be
 
-# Update
+OCM gives you a **local control panel** for the things that become painful first:
-git pull --ff-only
+- seeing your current setup clearly
-```
+- operating it more safely
+- recovering faster when something breaks
 
-Open [http://localhost:3333](http://localhost:3333) in your browser. For remote access, use `bash start.sh --host 0.0.0.0`.
+## Is OCM for you?
-## Docs
 
-- English guide: docs/USAGE_GUIDE.en.md
+**OCM is a good fit if:**
-- 中文使用说明: docs/USAGE_GUIDE.zh-CN.md
+- you already run OpenClaw locally
-- FAQ: docs/FAQ.md
+- you want a visual control panel for agents, routing, logs, health, and backups
-- Security notes: docs/SECURITY.md
+- you are tired of editing nested config by hand
-- Changelog: docs/CHANGELOG.md
+- you want safer rollback / recovery nearby
 
+**OCM is probably not the best starting point if:**
+- you have not installed OpenClaw yet
+- you expect every OS / channel workflow to be equally mature today
+- you want a hosted cloud dashboard instead of a local tool
 
-## Why OCM
+## What OCM helps you do
 
-OCM is best thought of as an **OpenClaw control panel**, not just a config editor.
+### 1) See your OpenClaw setup clearly
+- inspect agents and sub-agents in a tree
+- verify bindings and routing
+- view models, auth, usage, and health in one place
 
-It helps with three things that become painful fast in raw JSON / terminal workflows:
+### 2) Operate OpenClaw with less risk
+- restart gateway
+- inspect logs
+- run OpenClaw commands from the built-in terminal
+- make config changes with rollback nearby
 
-- **Agent topology** — see which agent is bound to which chat/channel/thread
+### 3) Recover faster when things go wrong
-- **Everyday operations** — run CLI commands, inspect health/logs, restart gateway
+- browse backups / snapshots
-- **Safer changes** — update config with backups and rollback nearby
+- roll back config safely
+- keep troubleshooting and operational tools close to the same UI
 
-## Features
+## See it quickly
+
+These are the highest-value surfaces for first-time users:
+
+<p>
+  <img src="docs/redacted-screenshots-refresh/dashboard-2026-03-07.jpg" width="860" />
+</p>
+
+<p>
+  <img src="docs/redacted-screenshots-refresh/agents-2026-03-07.jpg" width="280" />
+  <img src="docs/redacted-screenshots-refresh/cli-2026-03-07.jpg" width="280" />
+  <img src="docs/redacted-screenshots/actions.jpg" width="180" />
+</p>
+
+More screenshots: see the [full usage guide](docs/USAGE_GUIDE.en.md) or the gallery below.
 
 <details>
-<summary><b>Screenshots</b> (redacted: no personal paths, no Telegram IDs)</summary>
+<summary><b>Full screenshots gallery</b> (redacted: no personal paths, no Telegram IDs)</summary>
 
 <p>
   <img src="docs/redacted-screenshots/dashboard.jpg" width="240" />
@@ -62,7 +86,77 @@ It helps with three things that become painful fast in raw JSON / terminal workf
 
 </details>
 
-- **Agent Management** — Add main agents and sub-agents through a guided setup flow. View all agents in a tree structure with model selection, workspace browsing, and inline configuration.
+## Quick Start (recommended path)
+
+```bash
+# Clone
+git clone https://github.com/dtzp555-max/ocm.git
+cd ocm
+
+# Run (no npm install needed)
+bash start.sh        # macOS / Linux
+start.bat            # Windows
+```
+
+Then open [http://localhost:3333](http://localhost:3333).
+
+For remote access, use:
+
+```bash
+bash start.sh --host 0.0.0.0
+```
+
+## First-run checklist
+
+For a good first experience, do these first:
+
+1. Confirm OCM found the correct OpenClaw directory
+2. Open **Dashboard** and check gateway health
+3. Open **Agents** / **Channels** and inspect bindings
+4. Open the built-in **Terminal** and run one OpenClaw command
+5. Check **Actions** / backup / rollback so recovery is nearby before you need it
+
+## Current support status
+
+**Best supported today**
+- macOS / Linux
+- local OpenClaw installations
+- local health / logs / ops / backup workflows
+- Telegram-first agent management workflows
+
+**Supported, but not equally smooth in every case**
+- Discord agent / thread workflows
+- more advanced routing setups
+- mixed environment / service layouts
+
+**Not the primary path today**
+- fully beginner-first OpenClaw onboarding
+- every OS / shell / service combination
+- WhatsApp-heavy sub-agent topology workflows
+
+## Common friction points
+
+The most common reasons OCM feels rough are not usually “the UI is broken”, but environment mismatch:
+
+- OpenClaw itself is not fully healthy yet
+- wrong OpenClaw directory
+- path / permission issues
+- different service/runtime layouts across machines
+- channel workflow differences
+
+That is why OCM should be thought of as a **control panel for an existing OpenClaw setup**, not a magic replacement for all installation complexity.
+
+## Docs
+
+- **Start here**: docs/USAGE_GUIDE.en.md
+- 中文使用说明: docs/USAGE_GUIDE.zh-CN.md
+- FAQ: docs/FAQ.md
+- Security notes: docs/SECURITY.md
+- Changelog: docs/CHANGELOG.md
+
+## Features by area
+
+- **Agent Management** — Add main agents and sub-agents through a guided setup flow. View all agents in a tree structure with model selection, workspace browsing, and inline configuration. New agents automatically get default Skills and Tool Groups, or you can customize them via the built-in Skills & Tools picker.
 - **Usage Statistics** — Real token usage data parsed directly from OpenClaw session files. Breakdown by model, agent, and day with a visual chart.
 - **Model & Auth** — Configure models from your registered provider list. Follow built-in guides for provider authentication setup.
 - **Built-in CLI** — Available from any page (top-right **Terminal** + bottom dock). Run OpenClaw commands with **real-time streaming output**, **Tab completion**, built-in presets, and your own saved favorites.
@@ -72,11 +166,13 @@ It helps with three things that become painful fast in raw JSON / terminal workf
 
 ## What's New (v0.9.x)
 
+- **Skills & Tools auto-config**: new agents/sub-agents automatically get default Skills (`memory-continuity`, `session-logs`) and Tool Groups (`fs`, `runtime`, `memory`, `sessions_spawn`, `subagents`). A collapsible Skills & Tools picker in the creation form lets you customize per agent.
 - **Discord support (v0.9.x)**: add agent/sub-agent flows for Discord (main agent binds channel; sub-agent binds thread).
-- **README screenshots gallery**: all screenshots are now grouped into one collapsible section to keep the homepage short.
+- **README screenshots gallery**: screenshots grouped into a collapsible section so the homepage stays shorter.
 - **Built-in CLI terminal**: run OpenClaw commands from any page with streaming output, presets, favorites, and Tab completion.
 - **Ops actions**: restart gateway, view logs, health check, backups (local + NAS), and cron management.
 - **Telegram-first workflow**: safer sub-agent setup flow + allowlist helper + warnings for group privacy.
+
 ## Channel Support
 
 OCM is a local UI for OpenClaw. Core features like agent trees, routing, models/auth, ops, backups, and the built-in CLI are channel-agnostic.
@@ -184,6 +280,12 @@ ocm/
 - **Frontend**: Vanilla HTML/CSS/JS embedded in a template literal — no build step, no framework
 - **Architecture**: Single-file, zero-dependency, runs anywhere Node.js runs
 
+## Related Projects
+
+- [clawkeeper](https://github.com/dtzp555-max/clawkeeper) — OpenClaw memory operations kit for doctor / provider switch / backup / restore workflows. *(archived)*
+- [execution-agent-planner](https://github.com/dtzp555-max/execution-agent-planner) — archived planning skill for deciding when work should stay with one execution agent or split across specialized workers.
+- [gh-pr-release-flow](https://github.com/dtzp555-max/gh-pr-release-flow) — skill for PR-first GitHub repos where release work should happen after merge.
+
 ## License
 
 MIT
@@ -192,7 +294,38 @@ MIT
 
 ## 中文说明
 
-OpenClaw Manager 是一个零依赖的本地 Web 管理界面，用于可视化管理 [OpenClaw](https://github.com/anthropics/openclaw) AI 智能体。所有代码在一个 `.js` 文件中，只需 Node.js 18+，不需要 `npm install`。
+OpenClaw Manager（OCM）是一个本地 OpenClaw 控制面板，用来更直观地管理 Agent、查看绑定关系、做健康检查、运行内置 CLI，并在改配置或排错时少踩坑。
+
+### OCM 解决什么问题
+
+直接使用 OpenClaw 很强，但日常维护很容易变成：
+- 改 `openclaw.json` 容易手滑
+- Agent / bot / 群 / 线程绑定关系很难一眼看清
+- 健康检查、日志、重启、回滚散落在多个终端命令里
+- 出问题时恢复链路不够顺手
+
+OCM 的定位不是替代 OpenClaw，而是给现有 OpenClaw 环境加一个**本地控制台**。
+
+### 适合谁
+
+**适合：**
+- 已经在本机跑通 OpenClaw 的用户
+- 想更直观地管理 Agent / Routing / Health / Logs / Backups 的用户
+- 不想频繁手改复杂 JSON 的用户
+
+**暂时不太适合：**
+- 还没装好 OpenClaw、希望 OCM 包办全部安装复杂度的新手
+- 默认期待所有平台 / 所有 channel 路线都同样成熟的用户
+- 想要云端托管产品而不是本地工具的用户
+
+### OCM 最有价值的三件事
+
+1. **看清你的 OpenClaw 结构**
+   - Agent 树、绑定关系、模型、认证、健康状态更直观
+2. **更安全地做运维操作**
+   - 重启 gateway、看日志、跑 CLI、改配置、回滚更顺手
+3. **出问题时恢复更快**
+   - 备份、回滚、排查入口都更近
 
 ### 快速开始
 
@@ -203,49 +336,33 @@ bash start.sh        # macOS / Linux
 start.bat            # Windows
 ```
 
-访问 [http://localhost:3333](http://localhost:3333)。远程访问请使用 `bash start.sh --host 0.0.0.0`。
+打开 [http://localhost:3333](http://localhost:3333)。
 
-### 功能
+### 第一次使用建议先做这几步
 
-- **Agent 管理** — 添加主 Agent 和子 Agent，树状结构查看，内联切换模型，浏览工作区文件
+1. 确认 OCM 找到了正确的 OpenClaw 目录
-- **使用统计** — 从 OpenClaw 会话文件解析真实 Token 用量，按模型/Agent/日期维度展示
+2. 打开 **Dashboard** 看 gateway 是否健康
-- **模型与认证** — 管理注册模型列表，内置 Provider 认证引导
+3. 打开 **Agents / Channels** 看绑定是否正确
-- **内置终端** — 实时流式输出，预设命令，收藏夹，Tab 补全
+4. 打开内置 **Terminal** 跑一条 OpenClaw 命令
-- **运维面板** — 重启网关、查看日志、健康检查、本地/NAS 备份、Cron 任务管理
+5. 看一眼 **Actions / 备份 / 回滚**，确保恢复路径就在手边
-- **双语界面** — 中英文一键切换
 
-### Telegram 场景与安全提示
+### 当前支持定位
 
-OCM 主要面向 **Telegram 场景**：
+**目前最好用的场景：**
+- macOS / Linux
+- 本地 OpenClaw 环境
+- 本地 health / logs / ops / backup 工作流
+- Telegram-first 的 Agent 管理方式
 
-- 通过群组绑定主 Agent，并在每条树下管理多个 Sub-Agent
+**支持，但不保证同样顺滑：**
-- 让每个 Agent 拥有独立 `workspace`、`SOUL.md`、`MEMORY.md`
+- Discord workflow
-- 通过可视化方式更新 `openclaw.json`，减少手动改配置风险
+- 更复杂的 routing / service 布局
-- 建议使用者已具备基础 OpenClaw CLI 操作经验（如 `onboard`、认证、查看网关日志）
+- 混合环境
 
-关键安全设置（务必确认）：
+### 文档
 
-- BotFather 里 **Allow Groups = ON**
+- 英文使用说明：docs/USAGE_GUIDE.en.md
-- BotFather 里 **Group Privacy = OFF**
+- 中文使用说明：docs/USAGE_GUIDE.zh-CN.md
-- 每个 Agent 群组只保留“你自己 + 对应 Agent/Sub-Agent”
+- FAQ：docs/FAQ.md
-- 不要邀请其他人进组（会带来安全和 API 费用风险）
+- 安全说明：docs/SECURITY.md
-
+- 更新记录：docs/CHANGELOG.md
-### 配置
-
-启动脚本自动检测 `~/.openclaw` 目录。如需指定其他路径：
-
-```bash
-bash start.sh --dir /path/to/.openclaw --port 8080
-```
-
-也可手动创建 `manager-config.json`：
-
-```json
-{ "dir": "~/.openclaw" }
-```
-
-
-## Discord (thread-first) notes
-
-- Recommended: keep top-level agents in dedicated channels, and bind sub-agents to **threads** (one thread per task).
-- To switch a Telegram group / Discord thread binding: go to **Routing** → remove the old binding → add a new binding with the new ID/link.

docs/CHANGELOG.md

@@ -2,6 +2,11 @@
 
 This is a user-facing summary derived from `DEVLOG.md`.
 
+## v0.9.4
+- **Skills & Tools auto-config**: New agents/sub-agents now automatically get default Skills (memory-continuity, agent-workflow, execution-agent-dispatch, session-logs) and Tool Groups (fs, runtime, memory, sessions_spawn, subagents).
+- Agent creation forms include a collapsible Skills & Tools picker with checkbox multi-select and quick-action buttons (Default / All / None).
+- All 4 creation endpoints (Telegram/Discord × Agent/Sub-Agent) support `skills` and `toolGroups` fields.
+
 ## v0.9.0
 - Agents: Telegram + Discord support in Add Agent / Add Sub-Agent wizard (main agent binds channel; sub-agent binds thread).
 - UI: top-center OCM version badge; improved Telegram/Discord setup guides.

docs/SCREENSHOT_REFRESH_BRIEF.md

@@ -0,0 +1,79 @@
+# OCM Screenshot Refresh Brief
+
+Goal: refresh only the highest-value README screenshots first, not every screenshot.
+
+## Redaction rules
+Always hide or blur:
+- Telegram / Discord IDs
+- bindings / peer IDs / thread IDs
+- local filesystem paths containing Tao's name
+- account labels if they expose personal identifiers
+- anything that looks like a secret/token
+
+## Priority 1 screenshots
+
+### 1) Dashboard hero
+Purpose:
+- show OCM as a real control panel
+- support the README first-run / health story
+
+Wanted state:
+- Dashboard page open
+- gateway looks healthy if possible
+- cards readable
+- top bar clean
+- no obvious warning/noise unless it helps credibility
+
+Frame guidance:
+- browser window wide enough to show full dashboard naturally
+- keep top navigation visible
+- avoid cropping too tightly
+
+### 2) Agents page
+Purpose:
+- show the core product value: visible agent topology / structure
+
+Wanted state:
+- main agent + several sub-agents visible
+- structure easy to understand at a glance
+- avoid visual clutter where possible
+- keep model selectors/buttons visible, but not the focus
+
+Frame guidance:
+- enough width to show the tree clearly
+- IDs and personal paths redacted
+
+### 3) CLI / Terminal page
+Purpose:
+- show that OCM is not only a viewer; it can operate OpenClaw too
+
+Wanted state:
+- built-in terminal open
+- a real command visible (recommended: `openclaw status` or `openclaw doctor`)
+- real output visible
+- do not use tutorial arrows / annotation bubbles
+
+Frame guidance:
+- include command input and some output
+- avoid empty terminal state
+- redact paths / personal names
+
+## Priority 2 screenshots
+- Actions menu
+- Backup / rollback
+- Channels
+- Models
+- Auth
+- Cron
+- Stats
+
+## Selected refresh set (2026-03-07)
+- Dashboard: `docs/redacted-screenshots-refresh/dashboard-2026-03-07.jpg`
+- Agents: `docs/redacted-screenshots-refresh/agents-2026-03-07.jpg`
+- CLI: `docs/redacted-screenshots-refresh/cli-2026-03-07.jpg`
+- Keep using existing `docs/redacted-screenshots/actions.jpg` for now
+
+## Notes
+- Existing `actions.jpg` is already strong and can stay for now.
+- The new Dashboard and Agents captures are preferred for README hero/overview use.
+- The new CLI capture is preferred because it shows a real command and real output.

docs/USAGE_GUIDE.en.md

@@ -138,8 +138,9 @@ Click `+ Add Agent` and fill:
 - workspace path (recommended: separate folder per agent)
 - model (dropdown is sourced from `openclaw models list`)
 - the BotFather token
+- **Skills & Tools** (optional): expand the collapsible picker at the bottom to customize. By default, new agents get: `memory-continuity`, `session-logs` as Skills, and `group:fs`, `group:runtime`, `group:memory`, `sessions_spawn`, `subagents` as Tool Groups. Use the quick-action buttons (Default / All / None) to adjust.
 
-OCM writes the agent + Telegram account/binding into `openclaw.json`.
+OCM writes the agent + Telegram account/binding + skills + tool groups into `openclaw.json`.
 
 ---
 
@@ -176,6 +177,7 @@ Send a message in the new group. In logs you should see something like:
 - paste the Group ID
 - set workspace (recommended: its own folder)
 - choose model
+- **Skills & Tools** (optional): same picker as Add Agent — defaults are pre-selected, expand to customize
 
 **Step 4 — Allowlist (optional but recommended)**
 

docs/USAGE_GUIDE.zh-CN.md

@@ -123,8 +123,9 @@ bash start.sh
 - 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
+- **Skills & Tools**（可选）：展开表单底部的折叠面板可自定义。默认预设：Skills 包含 `memory-continuity`、`session-logs`；Tool Groups 包含 `group:fs`、`group:runtime`、`group:memory`、`sessions_spawn`、`subagents`。提供「仅默认 / 全选 / 清空」快捷按钮。
 
-保存后，OCM 会把这颗主 Agent 写进 `openclaw.json`，并把 bot token 绑定到对应 `accountId`。
+保存后，OCM 会把主 Agent + skills + tools 写进 `openclaw.json`，并把 bot token 绑定到对应 `accountId`。
 
 ---
 
@@ -159,6 +160,7 @@ openclaw gateway logs --follow
 - 填 Group ID
 - workspace（建议单独目录）
 - 选择模型
+- **Skills & Tools**（可选）：和 Add Agent 相同的选择器，默认预设已勾选，展开可自定义
 
 #### Step 4：写 allowlist（推荐）
 如果你启用了 Telegram 的 allowlist（例如 `channels.telegram.allowFrom`），

openclaw-manager.js

@@ -24,7 +24,7 @@ const SCRIPT_DIR = __dirname;
 const MANAGER_CONFIG = path.join(SCRIPT_DIR, 'manager-config.json');
 let PORT = 3333;
 let HOST = '0.0.0.0';
-const APP_VERSION = '0.9.0';
+const APP_VERSION = '1.0.0-rc1';
 // --port 参数
 const portIdx = process.argv.indexOf('--port');
 if (portIdx !== -1 && process.argv[portIdx + 1]) PORT = parseInt(process.argv[portIdx + 1]) || 3333;
@@ -63,6 +63,17 @@ function refreshPaths() {
   CONFIG_PATH  = path.join(OPENCLAW_DIR, 'openclaw.json');
 }
 
+// ── Default Skills & Tool Groups for new agents ─────────────
+const DEFAULT_SKILLS = ['memory-continuity', 'session-logs'];
+const DEFAULT_TOOL_GROUPS = ['group:fs', 'group:runtime', 'group:memory', 'sessions_spawn', 'subagents'];
+
+function applySkillsTools(agentEntry, skills, toolGroups) {
+  const s = (Array.isArray(skills) && skills.length > 0) ? skills : DEFAULT_SKILLS;
+  const tg = (Array.isArray(toolGroups) && toolGroups.length > 0) ? toolGroups : DEFAULT_TOOL_GROUPS;
+  agentEntry.skills = s;
+  agentEntry.tools = { alsoAllow: tg };
+}
+
 // ── 已知模型列表 ──────────────────────────────────────────────
 const KNOWN_MODELS = [
   { id: '__default__',                              label: '使用全局默认模型' },
@@ -315,6 +326,28 @@ ${hasMemory ? initialMemory : '> 暂无初始记录。记忆将通过日常对
 `;
 }
 
+function generateCurrentStateMd() {
+  return `# CURRENT_STATE
+
+_Last updated: TBD Australia/Brisbane_
+
+## In Flight
+- none
+
+## Blocked / Waiting
+- none
+
+## Recently Finished
+- none
+
+## Next
+- none
+
+## Reset Summary
+- No active summary yet.
+`;
+}
+
 // ── 请求解析 ──────────────────────────────────────────────────
 function parseBody(req) {
   return new Promise((resolve, reject) => {
@@ -395,6 +428,12 @@ async function handleApi(req, res, urlObj, body) {
     );
     // For main agent without explicit binding, infer its accountId from first unclaimed telegram account
     const mainInferredAccountId = telegramAccounts.find(a => !claimedAccounts.has(a)) || telegramAccounts[0] || null;
+    // Build map: agentId -> parentAgentId inferred from subagents.allowAgents
+    const allowAgentsParentMap = {};
+    list.forEach(parent => {
+      const allowed = parent.subagents?.allowAgents || [];
+      allowed.forEach(childId => { if (!allowAgentsParentMap[childId]) allowAgentsParentMap[childId] = parent.id; });
+    });
     const enriched = list.map(a => {
       const binding = bindings.find(b => b.agentId === a.id && b.match?.peer?.kind === 'group');
       const groupId = binding?.match?.peer?.id || null;
@@ -424,7 +463,9 @@ async function handleApi(req, res, urlObj, body) {
       // Workspace: explicit per-agent, or defaults.workspace for main
       const workspace = a.workspace || (isMain ? defaultWorkspace : null);
       return { ...a, workspace, groupId, requireMention: groupId ? (groups[groupId]?.requireMention ?? true) : null,
-        effectiveModel: modelVal || defaults.model?.primary || '默认', hasOwnBot, accountId, parentAccountId, parentAgentId: a.parentAgentId || null, bindings: (bindings||[]).filter(b=>b.agentId===a.id).map(b=>({
+        effectiveModel: modelVal || defaults.model?.primary || '默认', hasOwnBot, accountId, parentAccountId,
+        parentAgentId: a.parentAgentId || (!hasOwnBot ? allowAgentsParentMap[a.id] : null) || null,
+        bindings: (bindings||[]).filter(b=>b.agentId===a.id).map(b=>({
         idx: bindings.indexOf(b),
         channel: b.match?.channel || '',
         accountId: b.match?.accountId || '',
@@ -439,7 +480,7 @@ async function handleApi(req, res, urlObj, body) {
 
   // POST /api/agents/bot — create agent with its own bot token
   if (method === 'POST' && pathname === '/api/agents/bot') {
-    const { botToken, agentId, name, model, workspace, purpose, personality } = body;
+    const { botToken, agentId, name, model, workspace, purpose, personality, skills, toolGroups } = body;
     if (!botToken || !botToken.trim()) {
       res.writeHead(400); res.end(JSON.stringify({ error: 'Bot Token is required' })); return;
     }
@@ -485,6 +526,7 @@ async function handleApi(req, res, urlObj, body) {
     const wsAlias = `~/.openclaw/workspaces/${workspace}`;
     const agentEntry = { id: agentId, name: name || agentId, workspace: wsAlias };
     if (model && model !== '__default__') agentEntry.model = { primary: model };
+    applySkillsTools(agentEntry, skills, toolGroups);
     cfg.agents.list.push(agentEntry);
     // Add binding
     if (!cfg.bindings) cfg.bindings = [];
@@ -497,6 +539,7 @@ async function handleApi(req, res, urlObj, body) {
     const agentName = name || agentId;
     await fsp.writeFile(path.join(wsPath, 'SOUL.md'), generateSoulMd(agentName, purpose || '', personality || ''), 'utf8');
     await fsp.writeFile(path.join(wsPath, 'MEMORY.md'), generateMemoryMd(agentName, ''), 'utf8');
+    await fsp.writeFile(path.join(wsPath, 'memory', 'CURRENT_STATE.md'), generateCurrentStateMd(), 'utf8');
     // Create agents/<id>/ runtime directory (required by OpenClaw gateway)
     const agentRuntimeDir = path.join(OPENCLAW_DIR, 'agents', agentId);
     await fsp.mkdir(agentRuntimeDir, { recursive: true });
@@ -506,7 +549,7 @@ async function handleApi(req, res, urlObj, body) {
       ok: true, agentId, workspacePath: wsPath, configBackup: bakPath,
       notes: [
         'Agent created with its own bot token',
-        'Workspace directory created with SOUL.md and MEMORY.md',
+        'Workspace directory created with SOUL.md, MEMORY.md, and memory/CURRENT_STATE.md',
         'Runtime directory created at agents/' + agentId + '/',
         'Configuration updated and backed up',
         'Restart gateway to load new bot: openclaw gateway restart'
@@ -517,7 +560,7 @@ async function handleApi(req, res, urlObj, body) {
 
   // POST /api/agents/discord — create top-level Discord agent (single bot, channel binding)
   if (method === 'POST' && pathname === '/api/agents/discord') {
-    const { agentId, name, workspaceFolder, model, purpose, personality, guildId, channelId } = body;
+    const { agentId, name, workspaceFolder, model, purpose, personality, guildId, channelId, skills, toolGroups } = body;
     if (!agentId || !/^[a-zA-Z0-9_-]+$/.test(agentId)) {
       res.writeHead(400); res.end(JSON.stringify({ error: 'Agent ID must contain only alphanumeric characters, underscores, or dashes' })); return;
     }
@@ -543,6 +586,7 @@ async function handleApi(req, res, urlObj, body) {
     const wsPath = path.join(OPENCLAW_DIR, 'workspaces', folder);
     const agentEntry = { id: agentId, name: name || agentId, workspace: wsAlias };
     if (model && model !== '__default__') agentEntry.model = { primary: model };
+    applySkillsTools(agentEntry, skills, toolGroups);
     cfg.agents.list.push(agentEntry);
 
     if (!cfg.bindings) cfg.bindings = [];
@@ -565,6 +609,7 @@ async function handleApi(req, res, urlObj, body) {
     await fsp.mkdir(path.join(wsPath, 'memory'), { recursive: true });
     await fsp.writeFile(path.join(wsPath, 'SOUL.md'), generateSoulMd(name || agentId, purpose || '', personality || ''), 'utf8');
     await fsp.writeFile(path.join(wsPath, 'MEMORY.md'), generateMemoryMd(name || agentId, ''), 'utf8');
+    await fsp.writeFile(path.join(wsPath, 'memory', 'CURRENT_STATE.md'), generateCurrentStateMd(), 'utf8');
 
     const agentRuntimeDir = path.join(OPENCLAW_DIR, 'agents', agentId);
     await fsp.mkdir(agentRuntimeDir, { recursive: true });
@@ -575,7 +620,7 @@ async function handleApi(req, res, urlObj, body) {
       notes: [
         'Discord agent created (single Discord bot)',
         'Channel binding added',
-        'Workspace + runtime directories created',
+        'Workspace + runtime directories created (including memory/CURRENT_STATE.md)',
         'Configuration updated and backed up',
         'Restart gateway to apply: openclaw gateway restart'
       ]
@@ -585,7 +630,7 @@ async function handleApi(req, res, urlObj, body) {
 
   // POST /api/agents/discord-sub — create Discord sub-agent (thread-only binding, grouping under parentAgentId)
   if (method === 'POST' && pathname === '/api/agents/discord-sub') {
-    const { agentId, displayName, workspaceFolder, model, purpose, personality, initialMemory, parentAgentId, guildId, threadId } = body;
+    const { agentId, displayName, workspaceFolder, model, purpose, personality, initialMemory, parentAgentId, guildId, threadId, skills, toolGroups } = body;
     if (!agentId || !/^[a-zA-Z0-9_-]+$/.test(agentId)) {
       res.writeHead(400); res.end(JSON.stringify({ error: 'Agent ID must contain only alphanumeric characters, underscores, or dashes' })); return;
     }
@@ -611,6 +656,7 @@ async function handleApi(req, res, urlObj, body) {
     const wsPath = path.join(OPENCLAW_DIR, 'workspaces', folder);
     const agentEntry = { id: agentId, name: displayName || agentId, workspace: wsAlias, parentAgentId: String(parentAgentId).trim() };
     if (model && model !== '__default__') agentEntry.model = { primary: model };
+    applySkillsTools(agentEntry, skills, toolGroups);
     cfg.agents.list.push(agentEntry);
 
     if (!cfg.bindings) cfg.bindings = [];
@@ -634,6 +680,7 @@ async function handleApi(req, res, urlObj, body) {
     const name = displayName || agentId;
     await fsp.writeFile(path.join(wsPath, 'SOUL.md'),   generateSoulMd(name, purpose || '', personality || ''), 'utf8');
     await fsp.writeFile(path.join(wsPath, 'MEMORY.md'), generateMemoryMd(name, initialMemory || ''), 'utf8');
+    await fsp.writeFile(path.join(wsPath, 'memory', 'CURRENT_STATE.md'), generateCurrentStateMd(), 'utf8');
 
     const agentRuntimeDir = path.join(OPENCLAW_DIR, 'agents', agentId);
     await fsp.mkdir(agentRuntimeDir, { recursive: true });
@@ -643,7 +690,7 @@ async function handleApi(req, res, urlObj, body) {
     res.end(JSON.stringify({ ok: true, agentId, workspacePath: wsPath, configBackup: bakPath,
       notes: [
         'Discord sub-agent created (thread-only binding)',
-        'Workspace + runtime directories created',
+        'Workspace + runtime directories created (including memory/CURRENT_STATE.md)',
         'Configuration updated and backed up',
         'Restart gateway to apply: openclaw gateway restart'
       ]
@@ -654,7 +701,7 @@ async function handleApi(req, res, urlObj, body) {
 
   // POST /api/agents — create sub-agent (shares parent bot)
   if (method === 'POST' && pathname === '/api/agents') {
-    const { agentId, displayName, groupId, workspaceFolder, model, purpose, personality, initialMemory, parentAgentId, telegramUserId } = body;
+    const { agentId, displayName, groupId, workspaceFolder, model, purpose, personality, initialMemory, parentAgentId, telegramUserId, skills, toolGroups } = body;
     if (!agentId || !/^[a-zA-Z0-9_-]+$/.test(agentId)) {
       res.writeHead(400); res.end(JSON.stringify({ error: 'Agent ID must contain only alphanumeric characters, underscores, or dashes' })); return;
     }
@@ -683,6 +730,7 @@ async function handleApi(req, res, urlObj, body) {
     const wsAlias= `~/.openclaw/workspaces/${folder}`;
     const agentEntry = { id: agentId, name: displayName || agentId, workspace: wsAlias, parentAgentId: String(parentAgentId).trim() };
     if (model && model !== '__default__') agentEntry.model = { primary: model };
+    applySkillsTools(agentEntry, skills, toolGroups);
     cfg.agents.list.push(agentEntry);
     // Binding uses parent's accountId
     const parentAccountId = parentBinding.match.accountId;
@@ -707,6 +755,7 @@ async function handleApi(req, res, urlObj, body) {
     const name = displayName || agentId;
     await fsp.writeFile(path.join(wsPath, 'SOUL.md'),   generateSoulMd(name, purpose, personality), 'utf8');
     await fsp.writeFile(path.join(wsPath, 'MEMORY.md'), generateMemoryMd(name, initialMemory), 'utf8');
+    await fsp.writeFile(path.join(wsPath, 'memory', 'CURRENT_STATE.md'), generateCurrentStateMd(), 'utf8');
     // Create agents/<id>/ runtime directory (required by OpenClaw gateway)
     const agentRuntimeDir = path.join(OPENCLAW_DIR, 'agents', agentId);
     await fsp.mkdir(agentRuntimeDir, { recursive: true });
@@ -715,7 +764,7 @@ async function handleApi(req, res, urlObj, body) {
     res.end(JSON.stringify({ ok: true, agentId, workspacePath: wsPath, configBackup: bakPath,
       notes: [
         'Sub-agent created and shares parent bot',
-        'Workspace and runtime directories created',
+        'Workspace and runtime directories created (including memory/CURRENT_STATE.md)',
         'Configuration updated and backed up',
         'Restart gateway to apply: openclaw gateway restart'
       ],
@@ -1700,6 +1749,19 @@ header { background:var(--surface); border-bottom:1px solid var(--border); paddi
 .lang-toggle { background:var(--border); border:none; color:var(--muted); border-radius:6px; padding:5px 10px; font-size:12px; cursor:pointer; }
 .lang-toggle:hover { background:#3a3f5c; color:var(--text); }
 
+/* skills/tools picker */
+.skills-picker { background:var(--bg); border:1px solid var(--border); border-radius:10px; padding:12px 14px; margin-top:4px; }
+.skills-picker summary { font-size:13px; font-weight:600; cursor:pointer; color:var(--text); user-select:none; }
+.skills-picker summary:hover { color:var(--accent); }
+.skills-picker .sp-grid { display:grid; grid-template-columns:1fr 1fr; gap:4px 16px; margin-top:8px; }
+.skills-picker label.sp-item { display:flex; align-items:center; gap:6px; font-size:12.5px; color:var(--muted); cursor:pointer; padding:3px 0; }
+.skills-picker label.sp-item:hover { color:var(--text); }
+.skills-picker label.sp-item input[type=checkbox] { accent-color:var(--accent); width:14px; height:14px; }
+.skills-picker .sp-actions { display:flex; gap:8px; margin-top:8px; padding-top:8px; border-top:1px solid var(--border); }
+.skills-picker .sp-actions button { font-size:11px; padding:3px 10px; border-radius:6px; border:1px solid var(--border); background:var(--surface); color:var(--muted); cursor:pointer; }
+.skills-picker .sp-actions button:hover { color:var(--text); border-color:var(--accent); }
+.sp-section-title { font-size:11px; font-weight:700; color:var(--accent); text-transform:uppercase; letter-spacing:.5px; margin:8px 0 4px; grid-column:1/-1; }
+
 /* centered version badge */
 .top-version{ position:absolute; left:50%; transform:translateX(-50%); font-size:12px; font-weight:800; letter-spacing:.3px; color:var(--text); background:rgba(108,99,255,.18); border:1px solid rgba(108,99,255,.45); padding:4px 12px; border-radius:999px; box-shadow:0 6px 18px rgba(0,0,0,.25); }
 .ver-old{ display:none; }
@@ -2030,6 +2092,16 @@ const MAIN_HTML_BODY = String.raw`
       </div>
     </div>
     <div class="dash-gauges" id="dashGauges"><div class="empty">Loading...</div></div>
+    <div class="card dash-notice">
+      <h3 data-i18n="dash.firstRunTitle">🚀 First-run checklist</h3>
+      <ul class="dash-note-list">
+        <li data-i18n="dash.firstRun1">Confirm OCM is pointed at the correct OpenClaw directory.</li>
+        <li data-i18n="dash.firstRun2">Check this Dashboard first: make sure Gateway is running and the system looks healthy.</li>
+        <li data-i18n="dash.firstRun3">Open Agents / Routing and confirm the bindings you expect are actually there.</li>
+        <li data-i18n="dash.firstRun4">Use the built-in Terminal to run <code>openclaw doctor</code> once so you catch environment/auth issues early.</li>
+        <li data-i18n="dash.firstRun5">Before bigger edits, make a backup or check rollback so recovery is nearby if something goes wrong.</li>
+      </ul>
+    </div>
     <div class="dash-sections">
       <div class="card dash-card" id="dashSystem">
         <h3>🖥️ System Info</h3>
@@ -2175,6 +2247,7 @@ const MAIN_HTML_BODY = String.raw`
       <select id="cliPreset" onchange="onCliPresetSelect()" style="font-size:11px;padding:3px 6px;max-width:180px;background:var(--bg);border:1px solid var(--border);border-radius:6px;color:var(--text)">
         <option value="" data-i18n="cli.presets">── 常用命令 ──</option>
       </select>
+      <button class="btn-secondary" style="font-size:11px;padding:3px 10px" onclick="copyCliCommand()" data-i18n="cli.copy">复制命令</button>
       <button class="btn-secondary" style="font-size:11px;padding:3px 10px" onclick="clearCliOutput()" data-i18n="cli.clear">清空</button>
       <button class="btn-secondary" style="font-size:11px;padding:3px 10px" onclick="toggleCliPanel()" data-i18n="cli.collapse">▼ 收起</button>
     </div>
@@ -2439,6 +2512,12 @@ const I18N = {
     'models.onlyCliHint':'模型下拉仅显示 openclaw models list 返回的模型。',
     'models.modelListErr':'读取 openclaw models list 失败：',
     'models.modelListEmpty':'未从 openclaw models list 解析到模型。请先运行 openclaw onboard 并确认模型可用。',
+    'dash.firstRunTitle':'🚀 首次使用检查清单',
+    'dash.firstRun1':'先确认 OCM 指向的是正确的 OpenClaw 数据目录。',
+    'dash.firstRun2':'先看 Dashboard：确认 Gateway 正在运行，系统状态看起来正常。',
+    'dash.firstRun3':'打开 Agents / Routing，确认你期望的绑定确实已经存在。',
+    'dash.firstRun4':'用内置终端先运行一次 <code>openclaw doctor</code>，尽早发现环境或认证问题。',
+    'dash.firstRun5':'在做较大改动前，先做一次备份或确认回滚入口，出问题时更容易恢复。',
     'dash.noticeTitle':'📌 Telegram 使用说明（重要）',
     'dash.notice1':'OCM 主要面向 Telegram：通过群组绑定 Agent，让每个 Agent 拥有独立 Workspace / SOUL.md / MEMORY.md。',
     'dash.notice2':'请确保你已有基本 OpenClaw 操作经验。OCM 主要负责可视化更新 openclaw.json，方便增删主 Agent 与 Sub-Agent，并支持多条 Agent 树。',
@@ -2521,6 +2600,9 @@ const I18N = {
     'wiz.soul':'性格关键词','wiz.soulHint':'（逗号分隔，选填）','wiz.soulPh':'幽默、直接、有条理...',
     'wiz.soulTip':'留空则使用默认成长型提示词（推荐）',
     'wiz.memory':'初始记忆','wiz.memoryHint':'（MEMORY.md，选填）','wiz.memoryPh':'例如：群组主要用中文交流。用户偏好简洁回复。',
+    'wiz.skillsTools':'Skills & Tools 配置','wiz.skillsHint':'不选择则使用默认预设（推荐）',
+    'wiz.skillsSection':'Skills（能力）','wiz.toolsSection':'Tool Groups（权限）',
+    'wiz.spSelectDefault':'仅默认','wiz.spSelectAll':'全选','wiz.spSelectNone':'清空',
     'wiz.preview':'即将创建：','wiz.group':'群组','wiz.modelLabel':'模型','wiz.globalDefault':'全局默认',
     'wiz.soulYes':'含性格关键词','wiz.soulDefault':'默认成长型提示词（推荐）',
     'wiz.autoBackup':'自动备份当前 openclaw.json ✓',
@@ -2534,7 +2616,7 @@ const I18N = {
     'ch.peerId':'Peer ID','ch.peerIdHint':'（群组 ID 或用户 ID）','ch.peerIdTip':'留空则匹配所有对应类型的 Peer',
     'ch.submit':'添加绑定',
     'l.sub':'选择运行模式',
-    'cli.open':'⌨️ 终端','cli.title':'⌨️ CLI 终端','cli.clear':'清空','cli.collapse':'▼ 收起',
+    'cli.open':'⌨️ 终端','cli.title':'⌨️ CLI 终端','cli.copy':'复制命令','cli.clear':'清空','cli.collapse':'▼ 收起',
     'cli.ready':'── 终端就绪，等待命令 ──','cli.cleared':'── 已清空 ──',
     'cli.presets':'── 常用命令 ──','cli.builtins':'内置命令','cli.favs':'我的收藏',
     'cli.run':'▶ 执行','cli.stop':'■ 停止','cli.star':'⭐','cli.manage':'管理',
@@ -2573,6 +2655,12 @@ const I18N = {
     'models.onlyCliHint':'Model dropdowns only show IDs returned by openclaw models list.',
     'models.modelListErr':'Failed to load openclaw models list: ',
     'models.modelListEmpty':'No model IDs were parsed from openclaw models list. Run openclaw onboard first.',
+    'dash.firstRunTitle':'🚀 First-run checklist',
+    'dash.firstRun1':'Confirm OCM is pointed at the correct OpenClaw directory.',
+    'dash.firstRun2':'Check Dashboard first: make sure Gateway is running and the system looks healthy.',
+    'dash.firstRun3':'Open Agents / Routing and confirm the bindings you expect are actually there.',
+    'dash.firstRun4':'Use the built-in Terminal to run <code>openclaw doctor</code> once so you catch environment or auth issues early.',
+    'dash.firstRun5':'Before bigger edits, make a backup or check rollback so recovery is nearby if something goes wrong.',
     'dash.noticeTitle':'📌 Telegram Usage Notes (Important)',
     'dash.notice1':'OCM is primarily for Telegram workflows: bind agents to groups so each agent has isolated Workspace / SOUL.md / MEMORY.md.',
     'dash.notice2':'Basic OpenClaw CLI experience is required. OCM focuses on visual openclaw.json updates for easier main-agent/sub-agent management and multiple agent trees.',
@@ -2656,6 +2744,9 @@ const I18N = {
     'wiz.soul':'Personality Keywords','wiz.soulHint':'(comma-separated, optional)','wiz.soulPh':'humorous, direct, organized...',
     'wiz.soulTip':'Leave blank for default growth prompt (recommended)',
     'wiz.memory':'Initial Memory','wiz.memoryHint':'(MEMORY.md, optional)','wiz.memoryPh':'e.g. Group mainly uses English. Users prefer concise replies.',
+    'wiz.skillsTools':'Skills & Tools','wiz.skillsHint':'Leave as-is for defaults (recommended)',
+    'wiz.skillsSection':'Skills','wiz.toolsSection':'Tool Groups',
+    'wiz.spSelectDefault':'Default','wiz.spSelectAll':'All','wiz.spSelectNone':'None',
     'wiz.preview':'About to create:','wiz.group':'Group','wiz.modelLabel':'Model','wiz.globalDefault':'Global Default',
     'wiz.soulYes':'With personality keywords','wiz.soulDefault':'Default growth prompt (recommended)',
     'wiz.autoBackup':'Auto-backup current openclaw.json ✓',
@@ -2669,7 +2760,7 @@ const I18N = {
     'ch.peerId':'Peer ID','ch.peerIdHint':'(Group ID or User ID)','ch.peerIdTip':'Leave blank to match all peers of this type',
     'ch.submit':'Add Binding',
     'l.sub':'Select Mode',
-    'cli.open':'⌨️ Terminal','cli.title':'⌨️ CLI Terminal','cli.clear':'Clear','cli.collapse':'▼ Collapse',
+    'cli.open':'⌨️ Terminal','cli.title':'⌨️ CLI Terminal','cli.copy':'Copy cmd','cli.clear':'Clear','cli.collapse':'▼ Collapse',
     'cli.ready':'── Terminal Ready ──','cli.cleared':'── Cleared ──',
     'cli.presets':'── Presets ──','cli.builtins':'Built-in','cli.favs':'My Favorites',
     'cli.run':'▶ Run','cli.stop':'■ Stop','cli.star':'⭐','cli.manage':'Manage',
@@ -2935,6 +3026,75 @@ function showAddForm(type) {
   applyLang();
 }
 
+// ── Skills / Tools picker helper ──────────────────────────
+const AVAILABLE_SKILLS = [
+  { id:'memory-continuity', label:'Memory Continuity', isDefault:true },
+  { id:'session-logs', label:'Session Logs', isDefault:true },
+  { id:'browser-use', label:'Browser Use', isDefault:false },
+  { id:'github', label:'GitHub', isDefault:false },
+  { id:'gh-issues', label:'GitHub Issues', isDefault:false },
+  { id:'coding-agent', label:'Coding Agent', isDefault:false },
+  { id:'discord', label:'Discord', isDefault:false },
+  { id:'weather', label:'Weather', isDefault:false },
+  { id:'summarize', label:'Summarize', isDefault:false },
+  { id:'healthcheck', label:'Healthcheck', isDefault:false },
+];
+const AVAILABLE_TOOL_GROUPS = [
+  { id:'group:fs', label:'Filesystem (group:fs)', isDefault:true },
+  { id:'group:runtime', label:'Runtime (group:runtime)', isDefault:true },
+  { id:'group:memory', label:'Memory (group:memory)', isDefault:true },
+  { id:'sessions_spawn', label:'Sessions Spawn', isDefault:true },
+  { id:'subagents', label:'Subagents', isDefault:true },
+];
+
+function buildSkillsToolsPicker(prefix) {
+  let html = '<details class="skills-picker" id="'+prefix+'-sp-wrap"><summary>' + t('wiz.skillsTools') + ' <span style="color:var(--muted);font-weight:400;font-size:12px">' + t('wiz.skillsHint') + '</span></summary>';
+  html += '<div class="sp-section-title">' + t('wiz.skillsSection') + '</div>';
+  html += '<div class="sp-grid">';
+  AVAILABLE_SKILLS.forEach(s => {
+    html += '<label class="sp-item"><input type="checkbox" data-sp-type="skill" value="'+s.id+'" '+(s.isDefault?'checked':'')+'>'+s.label+'</label>';
+  });
+  html += '</div>';
+  html += '<div class="sp-section-title">' + t('wiz.toolsSection') + '</div>';
+  html += '<div class="sp-grid">';
+  AVAILABLE_TOOL_GROUPS.forEach(g => {
+    html += '<label class="sp-item"><input type="checkbox" data-sp-type="tool" value="'+g.id+'" '+(g.isDefault?'checked':'')+'>'+g.label+'</label>';
+  });
+  html += '</div>';
+  html += '<div class="sp-actions">';
+  html += '<button type="button" onclick="spAction(\\x27'+prefix+'\\x27,\\x27default\\x27)">' + t('wiz.spSelectDefault') + '</button>';
+  html += '<button type="button" onclick="spAction(\\x27'+prefix+'\\x27,\\x27all\\x27)">' + t('wiz.spSelectAll') + '</button>';
+  html += '<button type="button" onclick="spAction(\\x27'+prefix+'\\x27,\\x27none\\x27)">' + t('wiz.spSelectNone') + '</button>';
+  html += '</div></details>';
+  return html;
+}
+
+function spAction(prefix, action) {
+  const wrap = document.getElementById(prefix+'-sp-wrap');
+  if (!wrap) return;
+  const boxes = wrap.querySelectorAll('input[type=checkbox]');
+  if (action==='none') { boxes.forEach(b=>b.checked=false); return; }
+  if (action==='all') { boxes.forEach(b=>b.checked=true); return; }
+  // default
+  const defSkills = AVAILABLE_SKILLS.filter(s=>s.isDefault).map(s=>s.id);
+  const defTools = AVAILABLE_TOOL_GROUPS.filter(g=>g.isDefault).map(g=>g.id);
+  boxes.forEach(b=>{
+    if (b.dataset.spType==='skill') b.checked = defSkills.includes(b.value);
+    else b.checked = defTools.includes(b.value);
+  });
+}
+
+function collectSkillsTools(prefix) {
+  const wrap = document.getElementById(prefix+'-sp-wrap');
+  if (!wrap) return { skills:[], toolGroups:[] };
+  const skills = []; const toolGroups = [];
+  wrap.querySelectorAll('input[type=checkbox]:checked').forEach(b=>{
+    if (b.dataset.spType==='skill') skills.push(b.value);
+    else toolGroups.push(b.value);
+  });
+  return { skills, toolGroups };
+}
+
 function buildAddAgentForm() {
   const modelOpts = buildModelOpts('__default__');
   return '<div class="add-form">' +
@@ -2992,6 +3152,7 @@ function buildAddAgentForm() {
     '<div class="form-group"><label>' + t('wiz.soul') + ' <span style="color:var(--muted);font-weight:400">' + t('wiz.soulHint') + '</span></label>' +
     '<input id="fa-soul" placeholder="' + t('wiz.soulPh') + '">' +
     '<span class="hint-text">' + t('wiz.soulTip') + '</span></div>' +
+    '<div class="form-group">' + buildSkillsToolsPicker('fa') + '</div>' +
     '<div style="display:flex;gap:8px;margin-top:14px">' +
     '<button class="btn-primary" onclick="submitAddAgent()">' + t('agents.addAgentSubmit') + '</button>' +
     '<button class="btn-ghost" onclick="clearAddForm()">' + t('btn.cancel') + '</button>' +
@@ -3087,6 +3248,7 @@ function buildAddSubForm() {
       '<input id="fs-soul" placeholder="' + t('wiz.soulPh') + '"></div>' +
     '<div class="form-group"><label>' + t('wiz.memory') + ' <span style="color:var(--muted);font-weight:400">' + t('wiz.memoryHint') + '</span></label>' +
       '<textarea id="fs-mem" placeholder="' + t('wiz.memoryPh') + '" rows="2"></textarea></div>' +
+    '<div class="form-group">' + buildSkillsToolsPicker('fs') + '</div>' +
 
     '<div style="display:flex;gap:8px;margin-top:14px">' +
       '<button class="btn-primary" onclick="submitAddSub()">' + t('agents.addSubSubmit') + '</button>' +
@@ -3120,18 +3282,20 @@ async function submitAddAgent() {
   if (!name) { toast(t('agents.errName'), 'err'); return; }
   if (!workspace) { toast('Workspace folder is required', 'err'); return; }
 
+  const { skills, toolGroups } = collectSkillsTools('fa');
+
   try {
     if (channel === 'telegram') {
       const token = document.getElementById('fa-token').value.trim();
       if (!token) { toast(t('agents.errToken'), 'err'); return; }
-      const payload = { botToken: token, agentId, name, workspace, model: model === '__default__' ? '' : model, purpose, personality };
+      const payload = { botToken: token, agentId, name, workspace, model: model === '__default__' ? '' : model, purpose, personality, skills, toolGroups };
       await api('POST', '/api/agents/bot', payload);
     } else {
       const link = (document.getElementById('fa-discord-link')?.value||'').trim();
       const parsed = parseDiscordLinkOrId(link);
       if (!parsed) { toast('Discord channel link/ID is required', 'err'); return; }
       const payload = { agentId, name, workspaceFolder: workspace, model: model === '__default__' ? '' : model, purpose, personality,
-        guildId: parsed.guildId, channelId: parsed.channelId };
+        guildId: parsed.guildId, channelId: parsed.channelId, skills, toolGroups };
       await api('POST', '/api/agents/discord', payload);
     }
     toast('Agent created successfully', 'ok');
@@ -3160,6 +3324,8 @@ async function submitAddSub() {
   if (!/^[a-zA-Z0-9_-]+$/.test(agentId)) { toast(t('wiz.errIdFormat'), 'err'); return; }
   if (!workspaceFolder) { toast('Workspace folder is required', 'err'); return; }
 
+  const { skills, toolGroups } = collectSkillsTools('fs');
+
   try {
     if (channel === 'telegram') {
       const groupId = document.getElementById('fs-gid').value.trim();
@@ -3167,14 +3333,14 @@ async function submitAddSub() {
       if (!groupId) { toast(t('wiz.errGroupId'), 'err'); return; }
       if (telegramUserId && !/^\d+$/.test(telegramUserId)) { toast('Telegram User ID must be a number', 'err'); return; }
       await api('POST', '/api/agents', { parentAgentId, agentId, displayName: displayName || agentId, groupId,
-        workspaceFolder, model: model === '__default__' ? '' : model, purpose, personality, initialMemory, telegramUserId });
+        workspaceFolder, model: model === '__default__' ? '' : model, purpose, personality, initialMemory, telegramUserId, skills, toolGroups });
     } else {
       const link = (document.getElementById('fs-discord-link')?.value||'').trim();
       const parsed = parseDiscordLinkOrId(link);
       if (!parsed) { toast('Discord thread link/ID is required', 'err'); return; }
       await api('POST', '/api/agents/discord-sub', { parentAgentId, agentId, displayName: displayName || agentId,
         workspaceFolder, model: model === '__default__' ? '' : model, purpose, personality, initialMemory,
-        guildId: parsed.guildId, threadId: parsed.channelId });
+        guildId: parsed.guildId, threadId: parsed.channelId, skills, toolGroups });
     }
     toast(t('wiz.created'), 'ok');
     clearAddForm();
@@ -3191,16 +3357,24 @@ function renderAgents() {
   // Build grouping: prefer explicit parentAgentId; fallback to telegram bot-root grouping.
   const byId = {}; (S.agents||[]).forEach(a=>{ byId[a.id]=a; });
 
-  // infer parentAgentId for legacy telegram sub-agents if missing
+  // infer parentAgentId for sub-agents if missing
   const accountToRootId = {};
   (S.agents||[]).forEach(a=>{ if(a.hasOwnBot && a.accountId) accountToRootId[a.accountId]=a.id; });
   (S.agents||[]).forEach(a=>{
-    if(!a.parentAgentId && !a.hasOwnBot && a.parentAccountId && accountToRootId[a.parentAccountId]){
+    if(a.parentAgentId || a.hasOwnBot || a.id === 'main') return; // already resolved or is a root
+    // Case 1: legacy telegram sub-agent with parentAccountId
+    if(a.parentAccountId && accountToRootId[a.parentAccountId]){
       a._inferParentAgentId = accountToRootId[a.parentAccountId];
+      return;
+    }
+    // Case 2: orphan agent (no own bot, no parentAgentId, no binding-based inference)
+    // These are likely dynamically created sub-agents — default to 'main'
+    if(byId['main']){
+      a._inferParentAgentId = 'main';
     }
   });
 
-  const roots=[]; const childrenByRoot={}; const orphans=[];
+  const roots=[]; const childrenByRoot={};
   (S.agents||[]).forEach(a=>{
     const pid = a.parentAgentId || a._inferParentAgentId || '';
     if(pid && byId[pid]){
@@ -3209,6 +3383,12 @@ function renderAgents() {
       roots.push(a);
     }
   });
+  // Sort roots: agents with own bot first (main always first among those), then orphans
+  roots.sort((a,b)=>{
+    if(a.id==='main') return -1; if(b.id==='main') return 1;
+    if(a.hasOwnBot && !b.hasOwnBot) return -1; if(!a.hasOwnBot && b.hasOwnBot) return 1;
+    return (a.name||a.id).localeCompare(b.name||b.id);
+  });
 
   function fmtBinding(b){
     const ch = (b.channel||'').toLowerCase();
@@ -3668,7 +3848,14 @@ async function deleteAuth(key){
 }
 
 function copyText(txt){
-  navigator.clipboard.writeText(txt).then(()=>toast('已复制','success')).catch(()=>toast('复制失败','error'));
+  navigator.clipboard.writeText(txt).then(()=>toast(lang==='en'?'Copied':'已复制','success')).catch(()=>toast(lang==='en'?'Copy failed':'复制失败','error'));
+}
+
+function copyCliCommand(){
+  const inp=document.getElementById('cliInput');
+  const cmd=(inp&&inp.value?inp.value:'').trim();
+  if(!cmd){ toast(lang==='en'?'Nothing to copy (CLI input is empty)':'没有可复制的命令（输入框为空）','error'); return; }
+  copyText(cmd);
 }
 
 
@@ -3908,8 +4095,8 @@ function runCli(){
   es.addEventListener('done',e=>{
     try{
       const d=JSON.parse(e.data);
-      if(d.code===0) cliAppend('\\n✅ 完成 (exit 0)\\n','cli-done-ok');
+      if(d.code===0) cliAppend('\\n✅ Done (exit 0)\\n','cli-done-ok');
-      else cliAppend('\\n❌ 退出码 '+d.code+'\\n','cli-done-err');
+      else cliAppend('\\n❌ Exit code '+d.code+'\\n','cli-done-err');
     }catch(_){}
     es.close(); cliEvt=null; setCliRunning(false);
   });

package.json

@@ -1,6 +1,6 @@
 {
   "name": "openclaw-manager",
-  "version": "0.9.0",
+  "version": "1.0.0-rc1",
   "description": "A local web UI for managing OpenClaw AI agents \u2014 no npm install required",
   "main": "openclaw-manager.js",
   "scripts": {