Skip to content

Commit a17cab6

Browse files
YumengBaoagentkit-git
YumengBao
authored and
agentkit-git
committed
feat:add security skill
添加安装clawSentry skill GitOrigin-RevId: 5c1ddf221a048a67129ccc2e0d67ca33684e40c6
1 parent 1fdabc3 commit a17cab6

File tree

4 files changed

+165
-0
lines changed

4 files changed

+165
-0
lines changed

‎skills/byted-security-clawsentry/LICENSE‎

Lines changed: 13 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,13 @@
1+
# Copyright (c) 2025 Beijing Volcano Engine Technology Co., Ltd. and/or its affiliates.
2+
#
3+
# Licensed under the Apache License, Version 2.0 (the "License");
4+
# you may not use this file except in compliance with the License.
5+
# You may obtain a copy of the License at
6+
#
7+
# http://www.apache.org/licenses/LICENSE-2.0
8+
#
9+
# Unless required by applicable law or agreed to in writing, software
10+
# distributed under the License is distributed on an "AS IS" BASIS,
11+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12+
# See the License for the specific language governing permissions and
13+
# limitations under the License.

‎skills/byted-security-clawsentry/README.md‎

Lines changed: 78 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,78 @@
1+
# ClawSentry Readme
2+
3+
ClawSentry 是火山引擎 AI Assistant Security 为 OpenClaw 深度定制的安全防护组件。在保障 OpenClaw 正常作业的同时，提供实时风险监测与防御，涵盖提示词注入、敏感数据泄露及高危命令执行等核心安全场景。
4+
5+
## Skill 防护功能
6+
7+
- **提示词注入 (Prompt Injection) --- 识别恶意指令并拒绝执行**
8+
- 攻击者可能在你让 OpenClaw 处理的网页、文档或邮件中隐藏恶意指令，诱骗其执行非预期操作，如窃取密码或泄露数据。
9+
10+
- **敏感信息泄露 (PII Leakage) --- 防止敏感信息被直接读取、外发**
11+
- 你在与 AI 对话或处理文档时，可能无意间输入了个人或公司的敏感信息（如身份证号、手机号、API Key），这些信息可能被发送到外部大模型，造成隐私泄露。
12+
13+
- **高危操作 (Risky Operation) --- 拦截 AI 自动执行高权限风险操作**
14+
- AI 在理解模糊指令时可能出错，或被恶意诱导，调用高权限工具执行危险命令，例如将“清理桌面”错误地理解为删除系统重要文件。
15+
16+
- **恶意 Skill (Malicious Skill Attack) --- 扫描集成 Skill 的风险情况，阻止恶意 Skill 入侵**
17+
- 攻击者可能通过伪装成正常功能的第三方插件或技能（Skills），诱导用户安装。这些恶意技能在被 AI 调用时，可能在后台静默执行恶意代码，例如窃取浏览器凭证（Cookie）、抓取本地敏感文件或进行越权操作。
18+
19+
## 安装方式（仅需极简三步操作）
20+
21+
#### 通过对话直接安装
22+
23+
支持通过对话交互的方式完成 ClawSentry skill 包安装，例如你可以通过与 OpenClaw 对话完成安装。
24+
25+
#### 前提条件
26+
27+
在开始之前，请确保你已经正确安装了 OpenClaw 。
28+
29+
#### **步骤一 下载并运行 ClawSentry Skill**
30+
31+
**在对话框中输入：**帮我安装 byted-security-clawsentry
32+
33+
**预期结果：**执行完成后，在对话框中会返回安装成功的提示并提供登录授权链接。
34+
35+
#### 步骤二 授权与自动配置
36+
37+
1. 在打开的页面上，请仔细阅读相关服务协议和隐私政策。勾选同意后，单击 **“同意并开通”**。
38+
39+
2. 按提示完成登录，浏览器页面显示“授权成功”即表示完成。
40+
41+
42+
完成授权后，OpenClaw 会自动执行剩余安装和配置流程。
43+
44+
#### 步骤三 验证安装结果
45+
46+
完成安装后，你可以向你的 OpenClaw 通过对话框发送以下类似指令，来验证 ClawSentry 是否已成功安装并启用：
47+
48+
1. 查看 ClawSentry 的启用状态和版本号；
49+
2. 查看防护状态。
50+
51+
**预期结果**：对话框返回的信息中应包含 ClawSentry 插件，其版本为 **1.1.0**，且状态为**“已启用”**。
52+
53+
## 产品建议与反馈
54+
55+
安装或使用过程中，如果遇到任何问题或阻塞，你可以通过官网咨询直接与我们取得联系；如果想反馈产品建议，你可通过问卷提交相关内容。感谢你的支持与配合。
56+
57+
**官网咨询入口**
58+
59+
访问[火山引擎官网](https://www.volcengine.com/)，单击右侧**在线咨询**，进入对话。
60+
61+
**问卷表单**
62+
63+
访问链接填写问卷，反馈你的问题。
64+
65+
https://bytedance.larkoffice.com/share/base/form/shrcngOInnpkzC7OyN1y7QcwQJ8
66+
67+
**飞书群聊**
68+
69+
访问链接加入官方飞书群聊，参与产品讨论，获取产品最新动态。
70+
71+
https://applink.larkoffice.com/client/chat/chatter/add_by_link?link_token=845sa75d-18de-4b53-a623-42c4db4b25de
72+
73+
## 费用说明
74+
75+
当前产品处于**免费试用阶段**，暂不收取费用。
76+
77+
- 待产品进入正式收费阶段后，免费试用服务将相应停止。我们将提前发布收费标准及生效时间说明，并预留合理过渡期。
78+
- 在你主动确认开通付费服务前，不会产生任何费用。

‎skills/byted-security-clawsentry/SKILL.md‎

Lines changed: 63 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,63 @@
1+
---
2+
description: Install and configure the security-related plugins required by OpenClaw, including the `ai-assistant-security-openclaw` plugins. Use this skill when you want to complete installation and basic configuration of these plugins for an OpenClaw environment in one go.
3+
name: byted-security-clawsentry
4+
---
5+
6+
# ClawSentry
7+
8+
## Agent Rules
9+
10+
IMPORTANT: You MUST follow these steps EXACTLY when using this skill:
11+
12+
1. FIRST: Verify Node.js is installed on the system (version 14 or higher).
13+
2. SECOND: Execute the script `node {baseDir}/scripts/bundle.js`.
14+
3. THIRD: Verify the script executed successfully by checking:
15+
- Check the script exit code: 0 means success, non-0 means failure
16+
- Look for success output: "Login URL generated successfully:" in the script output
17+
- Look for the line that starts with `[LOGIN_URL]:` in the script output
18+
- Check if `{baseDir}/.state/login_state.json` file exists and contains a valid `loginUrl` field
19+
- If the file doesn't exist, wait a few seconds and try again
20+
- If you can't find the files, try looking in `.openclaw/skills/ClawSentry/scripts/` directory
21+
22+
4. FOURTH: After confirming the script succeeded, get the login URL using ONE of these methods (in this priority order):
23+
- Extract from the script output by looking for the line that starts with `[LOGIN_URL]:`
24+
- Read from `{baseDir}/.state/login_state.json` file (look for the `loginUrl` field)
25+
26+
5. FIFTH: Send the login URL to the user along with this message: "After completing login, please send me a message saying you have successfully logged in (e.g., '我已经注册成功了', '我已经登录成功了', '我已经注册好了', '登录好了', '登录上了')."
27+
6. SIXTH: After sending the login URL to the user, you MUST perform the following monitoring for 10 minutes (600 seconds):
28+
a. EVERY MINUTE, check if the process `bundle.js --login-token` is running using the command: `ps aux | grep "bundle.js --login-token"`
29+
b. If the process IS running: continue monitoring until the 10 minutes expire
30+
c. If the process IS NOT running:
31+
i. Read `{baseDir}/.state/login_state.json` file and check the `enable` field
32+
ii. If `enable` is `true`: END the conversation immediately
33+
iii. If `enable` is `false`: - Read `loginToken` and `deviceFingerprint` from `{baseDir}/.state/login_state.json` - Restart the process with: `node {baseDir}/scripts/bundle.js --login-token [loginToken] --device-fingerprint [deviceFingerprint]`
34+
d. Stop monitoring after 10 minutes have elapsed
35+
36+
DO NOT skip any steps. DO NOT modify the order of steps. ALWAYS verify the script succeeded before proceeding.
37+
38+
## Use Cases
39+
40+
- Use this skill when you need to install and prepare the `ai-assistant-security-openclaw` plugins for the OpenClaw environment on the current machine.
41+
- Trigger this skill when the user mentions requirements such as "install clawSentry / ai-assistant-security-openclaw plugin / install byted-security-clawsentry / 使用byted-security-clawsentry技能安装安全插件".
42+
43+
## Steps
44+
45+
### 1. Verify Node.js is installed:
46+
47+
Ensure Node.js (version 14 or higher) is installed on the system. You can check with:
48+
49+
```bash
50+
node --version
51+
```
52+
53+
### 2. Execute the combined script to install plugin and generate login link:
54+
55+
```bash
56+
node {baseDir}/scripts/bundle.js
57+
```
58+
59+
After the script completes:
60+
61+
- The script will display the login URL directly. You can also extract it from the output by looking for the line that starts with `[LOGIN_URL]:`
62+
- You can also read the login URL from `{baseDir}/.state/login_state.json` file (look for the `loginUrl` field). If the file doesn't exist, try reading it again after a few seconds. If you can't find the files, try looking in `.openclaw/skills/ClawSentry/scripts/` directory.
63+
- Send the login URL to the user along with this message: "After completing login, please send me a message saying you have successfully logged in (e.g., '我已经注册成功了', '我已经登录成功了', '我已经注册好了', '登录好了', '登录上了')."

0 commit comments

Comments
 (0)