← Back to Skills Marketplace
43
Downloads
0
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install chinese-axioma-guard
Description
数字免疫系统 - 保护 OpenClaw 代理免受恶意技能侵害。使用时机：(1) 安装新技能前，(2) 检测到可疑行为，(3) 扫描恶意软件模式，(4) 安全审计。本技能使用 Clawdex（来自 Koi）进行安全检查，并运行 4 个 Python 疫苗（C2、rootkits、chains、bootkits）扫描。
README (SKILL.md)
🛡️ Axiomata Guard — 数字免疫系统
保护 OpenClaw 代理免受恶意技能侵害
| 信息 | 值 |
|---|---|
| 版本 | 1.0.0 — 2026-05-07 |
| 状态 | 运行中 |
1. 目的和范围
目标
为 OpenClaw 代理提供数字免疫系统，检测和阻止恶意技能。
使用时机
| 触发器 | 行动 |
|---|---|
| 安装新技能前 | 运行 Clawdex 安全检查 |
| 检测到可疑行为 | 扫描恶意软件模式 |
| 安全审计 | 运行完整扫描 |
| "scan skill" | 执行安全扫描 |
2. 安全检查协议
第一层：Clawdex 检查
在安装任何技能之前：
1. 使用 Clawdex API 验证技能安全
2. 检查技能来源
3. 验证签名
第二层：Python 疫苗扫描
运行 4 个安全疫苗：
| 疫苗 | 检测内容 | 优先级 |
|---|---|---|
| C2 疫苗 | 命令控制通道 | 高 |
| Rootkit 疫苗 | Rootkit 模式 | 高 |
| Chains 疫苗 | 恶意链接 | 中 |
| Bootkit 疫苗 | 启动攻击 | 高 |
3. 扫描命令
扫描技能目录
# 运行完整安全扫描
python3 merlin-guard.py scan \x3Cskill-path>
# 仅 Clawdex 检查
python3 merlin-guard.py clawdex \x3Cskill-slug>
# 运行所有疫苗
python3 merlin-guard.py vaccines \x3Cskill-path>
快速检查
# 检查技能安全
clawhub inspect \x3Cskill-slug>
# 验证来源
curl -s https://clawhub.ai/api/skills/\x3Cslug>/verify
4. 威胁矩阵
| 威胁类型 | 描述 | 检测方法 |
|---|---|---|
| C2 | 命令控制通道 | 异常网络连接 |
| Rootkit | 系统隐藏进程 | 异常系统调用 |
| Chains | 恶意链接链 | 可疑 URL 模式 |
| Bootkit | 启动时攻击 | 修改启动脚本 |
5. 响应协议
检测到威胁
1. 隔离技能（不安装）
2. 记录到 memory.md
3. 通知 Alexandre
4. 提供威胁报告
安全通过
1. 允许安装
2. 记录到日志
3. 监控行为
6. 边缘情况
| 情况 | 处理方法 |
|---|---|
| Clawdex 无响应 | 使用本地疫苗作为后备 |
| 疫苗超时 | 跳过该疫苗，报告部分扫描 |
| 误报 | 记录并允许用户覆盖 |
In Altum Per Security. 🛡️ Axiomata Guard v1.0
Usage Guidance
Do not rely on this as a complete security gate until the maintainer aligns the documentation and code, implements real scan checks, verifies service status accurately, and removes or scopes the “notify Alexandre” workflow. If used, treat outputs as a limited reputation check and review any data sent to external or local services.
Capability Analysis
Type: OpenClaw Skill
Name: chinese-axioma-guard
Version: 1.0.2
The skill functions as a security scanner ('Digital Immune System') designed to verify the safety of other OpenClaw skills. The Python script `clawguard.py` performs lookups against an external API (clawdex.koi.security) and a local service to identify potential threats. There is no evidence of data exfiltration, malicious execution, or persistence; the network activity is limited to checking skill names against a reputation database, which is consistent with its stated purpose. A minor naming discrepancy exists between the documentation (referencing `merlin-guard.py`) and the actual filename (`clawguard.py`), but this appears to be a non-malicious inconsistency.
Capability Assessment
Purpose & Capability
The stated purpose as a security guard is coherent, but the artifacts do not support the advertised level of protection: the code mainly checks skill names against Clawdex and can report services as connected without verifying them.
Instruction Scope
The response protocol includes broad actions such as notifying “Alexandre,” recording to memory/logs, and monitoring behavior without defining recipients, scope, retention, or user approval.
Install Mechanism
There is no install spec, and the SKILL.md examples reference merlin-guard.py while the included file is clawguard.py, creating a setup/provenance mismatch.
Credentials
The code contacts Clawdex and a localhost Merlin API and reads local ./skills directory names; this is broadly aligned with a guard skill but should be clear to users.
Persistence & Privilege
No credentials, privileged writes, or destructive actions are shown, but the instructions mention memory/log persistence and ongoing monitoring without boundaries.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install chinese-axioma-guard - After installation, invoke the skill by name or use
/chinese-axioma-guard - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.2
中文版本首次发布
v1.0.1
中文版本首次发布
v1.0.0
中文版本首次发布
Metadata
Frequently Asked Questions
What is 中文 Axioma 守护?
数字免疫系统 - 保护 OpenClaw 代理免受恶意技能侵害。使用时机：(1) 安装新技能前，(2) 检测到可疑行为，(3) 扫描恶意软件模式，(4) 安全审计。本技能使用 Clawdex（来自 Koi）进行安全检查，并运行 4 个 Python 疫苗（C2、rootkits、chains、bootkits）扫描。 It is an AI Agent Skill for Claude Code / OpenClaw, with 43 downloads so far.
How do I install 中文 Axioma 守护?
Run "/install chinese-axioma-guard" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is 中文 Axioma 守护 free?
Yes, 中文 Axioma 守护 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does 中文 Axioma 守护 support?
中文 Axioma 守护 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created 中文 Axioma 守护?
It is built and maintained by Kofna3369 (@kofna3369); the current version is v1.0.2.
More Skills