senseguard

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

SenseGuard

Scans OpenClaw skills for natural language security threats — the kind that VirusTotal and traditional code scanners cannot detect.

Traditional scanners see SKILL.md as a text file. SenseGuard sees it as agent instructions and checks for prompt injection, data exfiltration, obfuscation, and persistence attacks.

扫描OpenClaw技能中的自然语言安全威胁——这类威胁是VirusTotal和传统代码扫描器无法检测到的。

传统扫描器将SKILL.md视为普通文本文件，而SenseGuard将其视为Agent指令，并检查是否存在prompt injection、data exfiltration、混淆和持久化攻击。

How to Use

使用方法

When the user asks to scan a skill:

bash

python3 scripts/scanner.py --target <skill-name-or-path>

Options:

```
--target all
```
— scan all installed skills
```
--deep
```
— force LLM semantic analysis (Layer 2)
```
--no-cache
```
— skip cached results
```
--json
```
— output raw JSON for further processing

当用户要求扫描技能时：

bash

python3 scripts/scanner.py --target <skill-name-or-path>

可选参数：

```
--target all
```
— 扫描所有已安装的技能
```
--deep
```
— 强制启用LLM语义分析（第二层）
```
--no-cache
```
— 跳过缓存结果
```
--json
```
— 输出原始JSON以进行后续处理

Layer 2 (Semantic Analysis)

第二层（语义分析）

When

--json

output contains a

layer2_prompt

field, process it as a security audit task, then feed the JSON result back to generate the final score. This is how the LLM analyzes intent beyond regex patterns.

当

--json

输出包含

layer2_prompt

字段时，将其作为安全审计任务处理，然后将JSON结果反馈回来以生成最终评分。这就是LLM超越正则表达式模式分析意图的方式。

Output

输出结果

The scanner outputs a Markdown risk report with:

Score (0-100) and rating: SAFE / CAUTION / DANGEROUS / MALICIOUS
Findings with rule IDs, evidence text, and line numbers
Actionable recommendations

For CRITICAL findings, clearly advise the user to take action.

扫描器会输出一份Markdown格式的风险报告，包含：

评分（0-100）和评级：SAFE / CAUTION / DANGEROUS / MALICIOUS
带有规则ID、证据文本和行号的检测结果
可执行的建议

对于CRITICAL检测结果，需明确建议用户采取行动。

Key Differentiator

核心优势

SenseGuard catches what VirusTotal cannot:

```
"ignore all previous instructions"
```
— prompt injection
```
curl -X POST
```
hidden in Markdown — data exfiltration
Zero-width characters hiding commands — obfuscation
```
"modify MEMORY.md"
```
— persistence attacks

These are invisible to traditional malware scanners because they target the AI agent, not the operating system.

SenseGuard能够捕捉VirusTotal无法检测到的威胁：

```
"ignore all previous instructions"
```
— prompt injection
隐藏在Markdown中的
```
curl -X POST
```
— data exfiltration
隐藏命令的零宽字符 — 混淆
```
"modify MEMORY.md"
```
— 持久化攻击

这些威胁对传统恶意软件扫描器来说是不可见的，因为它们针对的是AI Agent而非操作系统。