360 Secure SkillHub
360 安全技能中心

Skillguard

启动本地服务后，一次调用即可完成三项检查：恶意软件扫描、权限范围分析与提示注入检测，返回总体判定 SAFE/CAUTION/DANGEROUS、问题总数以及 scan、scope、injection 三个子报告。

mirni
mirni
热度 37.52
最近更新 2026-05-11 17:42:41
分类：安全依赖：代码执行应用场景：报告撰写

安装技能

复制下面的话术发给你的 Agent，它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill： npx secure-skills-cli add gh-skillguard
Metadata
name
gh-skillguard
description
Run a complete security audit on any OpenClaw SKILL.md in one call. Combines malware scanning (SkillScan), permission scope analysis (ScopeCheck), and prompt injection detection (PromptGuard) into a single unified report with a SAFE/CAUTION/DANGEROUS verdict.
metadata
{"openclaw":{"emoji":"🏰","requires":{"bins":["python"]},"install":[{"id":"pip","kind":"uv","packages":["fastapi","uvicorn","pydantic","pyyaml"]}]}}

SkillGuard

Full security audit of a SKILL.md — three checks, one call.

Start the server

uvicorn skillguard.app:app --port 8005

Audit a skill

curl -s -X POST http://localhost:8005/v1/audit-skill \
  -H "Content-Type: application/json" \
  -d "{\"skill_content\": $(cat path/to/SKILL.md | jq -Rs)}" | jq

Returns verdict (SAFE/CAUTION/DANGEROUS), total_findings, and three sub-reports:

  • scan — malware detection (safety_score, findings, verdict)
  • scope — permission analysis (declared vs detected, undeclared_access)
  • injection — prompt injection (risk_score, patterns_detected)

Why use SkillGuard instead of individual tools?

One call instead of three. Same price. Combined verdict logic: if the malware scan finds anything, it's DANGEROUS. If only scope or injection issues, it's CAUTION. Clean skill = SAFE.