d-run · windsonsea · Mar 13, 2026 · Mar 13, 2026
diff --git a/docs/zh/docs/en/clawos/faq.md b/docs/zh/docs/en/clawos/faq.md
@@ -0,0 +1,43 @@
+# FAQs
+
+## Will OpenClaw lose its memory if an instance has issues?
+
+**No**
+
+Whether you restart OpenClaw, pause the instance (feature coming soon), or release the instance, the memory will not be lost.
+
+This is because all data in `~/.openclaw` is persistently stored in the d.run storage system.
+
+## Is my OpenClaw secure?
+
+**Yes**
+
+DaoCloud’s ClawOS is an enterprise-grade “lobster farming platform” designed for security:
+
+- Uses container technology as the security sandbox for each OpenClaw instance  
+- Based on Kubernetes to ensure data security across compute, network, and storage layers  
+- Data and conversations are powered by the d.run large-model service platform, ensuring sensitive data protection  
+
+## Can I upload files for OpenClaw to analyze?
+
+**Of course**
+
+You can do this in the following ways:
+
+1. Send files to OpenClaw through Feishu chat for processing  
+2. Upload and manage files through the d.run file management interface  
+
+![Upload File](./images/upload-file.png)
+
+## OpenClaw instance creation failed
+
+Please confirm:
+
+1. Real-name verification has been completed  
+2. Your account balance is greater than the price of the selected resources  
+
+## Feishu integration not working
+
+A single Feishu application can only be connected to **one OpenClaw instance** at a time.
+
+If you have multiple OpenClaw instances, each instance must use a **separate App ID and App Secret** .
diff --git a/docs/zh/docs/en/clawos/feishu-integration.md b/docs/zh/docs/en/clawos/feishu-integration.md
@@ -0,0 +1,199 @@
+# Integrate Feishu
+
+This document explains how to integrate OpenClaw with Feishu so you can interact with OpenClaw directly through Feishu.
+
+## Create a Feishu App
+
+### Create an Internal Enterprise App
+
+1. Log in to the [Feishu Open Platform](https://open.feishu.cn/app).
+
+2. Click the **Create Internal App** button.
+
+### Configure App Information
+
+Set the App name, description, and icon, then click **Create** to complete the creation.
+
+### Add Bot Capability
+
+1. In the left navigation tree, select **App Capabilities → Add Capabilities** .
+2. Select the **Add by Capability** tab.
+3. Click **Add** on the **Bot** capability card.
+
+## Configure Permissions
+
+### Import Permission Configuration
+
+1. In the left navigation tree, select **Development Configuration → Permission Management** .
+
+2. Click **Batch Import/Export Permissions** .
+
+3. In the **Import** tab, paste the [permission configuration code](#permission-config).
+
+4. Click **Next: Confirm New Permissions** .
+
+5. In the popup window, confirm the permissions and click **Apply** .
+
+### Permission Configuration Code
+
+<details id="permission-config">
+<summary>Click to view the permission configuration code</summary>
+
+```json
+{
+  "scopes": {
+    "tenant": [
+      "contact:contact.base:readonly",
+      "docx:document:readonly",
+      "im:chat:read",
+      "im:chat:update",
+      "im:message.group_at_msg:readonly",
+      "im:message.p2p_msg:readonly",
+      "im:message.pins:read",
+      "im:message.pins:write_only",
+      "im:message.reactions:read",
+      "im:message.reactions:write_only",
+      "im:message:readonly",
+      "im:message:recall",
+      "im:message:send_as_bot",
+      "im:message:send_multi_users",
+      "im:message:send_sys_msg",
+      "im:message:update",
+      "im:resource",
+      "App:App:self_manage",
+      "cardkit:card:write",
+      "cardkit:card:read"
+    ],
+    "user": [
+      "contact:user.employee_id:readonly",
+      "offline_access",
+      "base:app:copy",
+      "base:field:create",
+      "base:field:delete",
+      "base:field:read",
+      "base:field:update",
+      "base:record:create",
+      "base:record:delete",
+      "base:record:retrieve",
+      "base:record:update",
+      "base:table:create",
+      "base:table:delete",
+      "base:table:read",
+      "base:table:update",
+      "base:view:read",
+      "base:view:write_only",
+      "base:app:create",
+      "base:app:update",
+      "base:app:read",
+      "sheets:spreadsheet.meta:read",
+      "sheets:spreadsheet:read",
+      "sheets:spreadsheet:create",
+      "sheets:spreadsheet:write_only",
+      "docs:document:export",
+      "docs:document.media:upload",
+      "board:whiteboard:node:create",
+      "board:whiteboard:node:read",
+      "calendar:calendar:read",
+      "calendar:calendar.event:create",
+      "calendar:calendar.event:delete",
+      "calendar:calendar.event:read",
+      "calendar:calendar.event:reply",
+      "calendar:calendar.event:update",
+      "calendar:calendar.free_busy:read",
+      "contact:contact.base:readonly",
+      "contact:user.base:readonly",
+      "contact:user:search",
+      "docs:document.comment:create",
+      "docs:document.comment:read",
+      "docs:document.comment:update",
+      "docs:document.media:download",
+      "docs:document:copy",
+      "docx:document:create",
+      "docx:document:readonly",
+      "docx:document:write_only",
+      "drive:drive.metadata:readonly",
+      "drive:file:download",
+      "drive:file:upload",
+      "im:chat.members:read",
+      "im:chat:read",
+      "im:message",
+      "im:message.group_msg:get_as_user",
+      "im:message.p2p_msg:get_as_user",
+      "im:message:readonly",
+      "search:docs:read",
+      "search:message",
+      "space:document:delete",
+      "space:document:move",
+      "space:document:retrieve",
+      "task:comment:read",
+      "task:comment:write",
+      "task:task:read",
+      "task:task:write",
+      "task:task:writeonly",
+      "task:tasklist:read",
+      "task:tasklist:write",
+      "wiki:node:copy",
+      "wiki:node:create",
+      "wiki:node:move",
+      "wiki:node:read",
+      "wiki:node:retrieve",
+      "wiki:space:read",
+      "wiki:space:retrieve",
+      "wiki:space:write_only"
+    ]
+  }
+}
+```
+
+</details>
+
+> For detailed explanations of these permissions, see the [Feishu API Scope List](https://open.larkoffice.com/document/server-docs/App-scope/scope-list).
+
+## Publish the App
+
+### Create a Version and Publish
+
+1. Click the **Create Version** button at the top.
+
+2. Configure the App version number, default capabilities, and release notes as needed.
+
+3. Click **Save** at the bottom of the page to create the version.
+
+4. Click **Confirm Publish** in the upper-right corner to publish the App.
+
+### Obtain Credentials
+
+1. In the left navigation tree, select **Basic Information → Credentials & Basic Info** .
+
+2. In the **App Credentials** section, obtain and record the **App ID** and **App Secret** .
+
+## Configure Event Subscriptions and Callback Long Connection
+
+!!! caution
+
+    For bot Apps that have not yet configured a long connection, you must start an OpenClaw instance **after obtaining the App ID and App Secret** before saving the event subscription and callback long-connection configuration.
+
+Refer to the [OpenClaw Quick Start](./quickstart.md) guide to start an OpenClaw instance.
+
+### Configure Event Subscription
+
+1. Click **Events & Callbacks** in the left navigation and set **Event Subscription** to **Long Connection**.
+
+2. Add a **Message Receive Event** (other events can be added as needed).
+
+### Configure Callback
+
+1. Click **Callback Configuration**.
+
+2. Set the configuration method to **Long Connection**.
+
+3. Add a callback configuration.
+
+### Publish the Version
+
+Publish the version and wait for approval.
+
+## Start Chatting
+
+Find your bot in Feishu and start chatting!
+
diff --git a/docs/zh/docs/en/clawos/index.md b/docs/zh/docs/en/clawos/index.md
@@ -0,0 +1,76 @@
+# OpenClaw Agent: From “Can Chat” to “Can Get Work Done”
+
+## What Is OpenClaw
+
+**OpenClaw** is a next-generation open-source AI agent runtime that represents a major leap in AI capability. Instead of merely answering questions passively, it can actively plan tasks, call tools, and execute actions autonomously—completing complex end-to-end work like a real digital employee.
+
+| Capability Dimension | Traditional AI Assistant | OpenClaw Agent |
+|---|---|---|
+| Interaction Mode | Passive Q&A, single response | Active planning, multi-step autonomous execution |
+| Capability Boundary | Text output only | Tool usage, OS operations, code execution |
+| Applicable Scenarios | Information lookup, simple generation | End-to-end automated workflows |
+
+**OpenClaw** provides five native capability layers:
+
+- **User Access Layer:** Supports chat conversations, CLI command line, IDE plugins, IM tools such as Feishu / WeChat / DingTalk, and Web APIs  
+- **Agent Core Engine:** Task planning, context management, workflow engine, sub-agent orchestration, multi-agent architecture  
+- **Tool Execution Layer:** File system access, terminal commands, browser control, API calls, custom skills, and third-party tool integration  
+- **Memory & Knowledge Layer:** Short-term session memory, long-term vector storage, knowledge base retrieval, cross-session context persistence  
+- **Model Integration Layer:** Compatible with Claude, GPT, local open-source models, enterprise private models, and the OpenAI API protocol  
+
+## Using OpenClaw on d.run
+
+**ClawOS v0.1** is now available on **d.run**, providing an enterprise-grade hosted runtime platform for OpenClaw. There is no need to configure environments or manually connect models. Your AI digital employee can be ready to work within **five minutes**.
+
+For detailed instructions, see the [Quick Start](./quickstart.md) guide.
+
+### Feishu Integration
+
+**ClawOS v0.1** natively supports seamless **Feishu** integration. When creating an instance, simply enable the **Integrate Feishu** option and provide your Feishu application's **App ID** and **App Secret**.
+
+Once configured, OpenClaw can directly send and receive messages, process files, and reply in group chats within Feishu—without switching interfaces.
+
+For detailed setup instructions, see the [Feishu Integration](./feishu-integration.md) guide.
+
+## Typical Use Cases
+
+### HR Resume Screening at Scale
+
+When facing dozens of PDF resumes with different formats, OpenClaw can automatically read them, extract key technical skills, score and rank candidates based on job requirements, and output a structured evaluation report. What previously took half a day can now be completed in minutes.
+
+!!! note
+
+    Resumes contain sensitive personal information (PII). A private deployment of d.run ensures that data never leaves the internal network and prevents resumes from being sent to external public cloud APIs.
+
+### Software Development Assistant
+
+After connecting to GitHub and configuring a multi-agent architecture (Main Agent + Research + Reviewer + Codex), OpenClaw can dive into source code repositories to find bugs, submit pull requests, enforce CI rules, and create a fully personalized automated development orchestrator.
+
+### Batch Document Review
+
+Encapsulate a full workflow—**read documents → extract key data → validate rules → generate approval feedback**—into a custom Skill using `createSkill`. Next time, you can process dozens of quarterly reports in bulk and export a summarized CSV with a single click, eliminating repetitive work.
+
+## Product Advantages
+
+| Advantage | Description |
+| --- | --- |
+| One-click activation, ready to use | Administrators do not need to manually configure environments, connect models, or set permissions. The system automatically creates sandboxes and injects tokens, allowing business users to start immediately without IT involvement. |
+| Secure sandbox isolation | Each OpenClaw instance runs independently in a containerized security sandbox using DaoCloud’s proprietary **zestU** kernel-level isolation technology. Even if an instance is compromised or fails, the impact remains strictly contained within the sandbox and cannot penetrate the internal network or modify host files. |
+| Persistent data, memory never lost | The entire `~/.openclaw` directory is persistently stored in the d.run storage system. Whether OpenClaw is restarted, paused, or the instance is released, conversation history, configurations, and memory remain intact. |
+| Secure and controllable model access | All model calls go through the d.run AI gateway and security policies instead of directly connecting to the public internet. Your data and conversations are processed entirely within the d.run platform’s model services, ensuring protection of sensitive information. |
+| Fully accessible backend | d.run provides two ways to access the OpenClaw backend: [SSH login and JupyterLab](quickstart.md#_4). Both support full CLI operations for advanced users who need debugging or customization capabilities. |
+
+## Features
+
+| Feature | Current Status |
+|---|---|
+| [One-click OpenClaw instance creation](./quickstart.md#openclaw) | Available |
+| [Seamless Feishu integration](./feishu-integration.md) | Available |
+| Secure container sandbox isolation | Available |
+| Persistent data storage | Available |
+| [SSH / JupyterLab backend access](./quickstart.md#__tabbed_1_1) | Available |
+| File management and uploads | Available |
+| Token usage analytics and cost control | Coming soon |
+| More large-model support | Coming soon |
+| Unified instance management (admin view) | Planned |
+| Audit tracking and operation replay | Planned |
diff --git a/docs/zh/docs/en/clawos/quickstart.md b/docs/zh/docs/en/clawos/quickstart.md
@@ -0,0 +1,66 @@
+# Quick Start
+
+This guide will help you quickly create and use an OpenClaw instance.
+
+## Prerequisites
+
+- A d.run platform account  
+- Completed real-name verification  
+- Sufficient account balance or vouchers  
+
+## Create an OpenClaw Instance
+
+1. Select the **ClawOS** module and click the **Create** button on the right.
+
+2. Set an English **Name** for the OpenClaw instance and click **Confirm** in the lower-right corner.
+
+3. (Optional) d.run OpenClaw supports seamless Feishu integration. When creating an instance, enable **Integrate Feishu** and enter the Feishu configuration information.
+
+> For details on how to obtain the Feishu configuration and complete the integration, refer to the [Feishu Integration](./feishu-integration.md) document.
+
+4. Wait for the instance creation to complete.
+
+## Access OpenClaw
+
+When the instance status shows **Running**:
+
+1. Click **Access Tools → OpenClaw** on the right
+2. Open the OpenClaw management page
+
+!!! note
+
+    Due to network conditions, it may take 1–2 minutes before the page becomes accessible.
+
+## Start Using OpenClaw
+
+Click **Proceed to the website** and start chatting in the conversation window.
+
+## Debug OpenClaw in the Backend
+
+d.run provides multiple ways to operate the backend. You can access it via **SSH** or **JupyterLab** in the web interface.
+
+=== "Method 1: SSH Login"
+
+    Use SSH to directly enter the OpenClaw secure sandbox.
+
+=== "Method 2: Web Access"
+
+    Access the backend through the web interface.
+
+### Command Line Operations
+
+Before operating the OpenClaw CLI, switch to the `node` user:
+
+```bash
+su node  # switch to the node user
+````
+
+View installed skills:
+
+```bash
+openclaw skills list
+```
+
+## OpenClaw FAQ
+
+See the [FAQ](./faq.md) document.