中文 EN

扫描报告

扫描新文件

✓ 可信 — 风险评分 5/100
上次扫描：1 天前 重新扫描
5 /100
Proactive Agent
AI agent architecture pattern guide with memory protocols, WAL, and security hardening
This is a documentation-based AI agent pattern guide with supplementary security audit utilities. No malicious behavior, credential theft, or hidden functionality detected.
技能名称Proactive Agent
分析耗时22.5s
引擎pi
✓
可以安装
This skill is safe to use. It provides legitimate agent architecture patterns and a security audit utility.
资源类型声明权限推断权限状态证据
文件系统 NONE NONE — No filesystem access declared or used
网络访问 NONE NONE — No network requests in code
命令执行 NONE NONE — scripts/security-audit.sh is read-only analysis, not execution
9 项发现
🔗
中危 外部 URL 外部 URL
https://img.shields.io/badge/quality-90%2B-brightgreen
README.md:3
🔗
中危 外部 URL 外部 URL
https://img.shields.io/badge/optimized-karpathy%20style-blue
README.md:4
🔗
中危 外部 URL 外部 URL
https://img.shields.io/badge/tests-✅-brightgreen
README.md:5
🔗
中危 外部 URL 外部 URL
https://img.shields.io/badge/examples-✅-brightgreen
README.md:6
🔗
中危 外部 URL 外部 URL
https://img.shields.io/badge/docs-✅-brightgreen
README.md:7
🔗
中危 外部 URL 外部 URL
https://img.shields.io/badge/batch-2026--03--16-orange
README.md:8
🔗
中危 外部 URL 外部 URL
https://x.com/halthelobster
SKILL-v2.3-backup.md:526
🔗
中危 外部 URL 外部 URL
https://clawdhub.com/halthelobster/bulletproof-memory
SKILL-v2.3-backup.md:550
🔗
中危 外部 URL 外部 URL
https://clawdhub.com/halthelobster/para-second-brain
SKILL-v2.3-backup.md:550

目录结构

19 文件 · 86.5 KB · 2897 行
Markdown 14f · 2616L Shell 1f · 149L Python 2f · 75L JSON 2f · 57L
├─ ▾ 📁 assets
│ ├─ 📝 AGENTS.md Markdown 155L · 3.8 KB
│ ├─ 📝 HEARTBEAT.md Markdown 128L · 2.9 KB
│ ├─ 📝 MEMORY.md Markdown 47L · 836 B
│ ├─ 📝 ONBOARDING.md Markdown 103L · 2.4 KB
│ ├─ 📝 SOUL.md Markdown 40L · 1.4 KB
│ ├─ 📝 TOOLS.md Markdown 55L · 1.0 KB
│ └─ 📝 USER.md Markdown 36L · 926 B
├─ ▾ 📁 references
│ ├─ 📝 onboarding-flow.md Markdown 158L · 4.0 KB
│ └─ 📝 security-patterns.md Markdown 109L · 3.0 KB
├─ ▾ 📁 scripts
│ ├─ 🐍 examples.py Python 35L · 912 B
│ ├─ 🔧 security-audit.sh Shell 149L · 4.0 KB
│ └─ 🐍 test.py Python 40L · 879 B
├─ 📋 _meta.json JSON 23L · 682 B
├─ 📋 clawhub.config.json JSON 34L · 694 B
├─ 📝 OPTIMIZATION.md Markdown 68L · 1.6 KB
├─ 📝 README.md Markdown 16L · 607 B
├─ 📝 SKILL-v2.3-backup.md Markdown 554L · 19.7 KB
├─ 📝 SKILL-v3-draft.md Markdown 499L · 16.6 KB
└─ 📝 SKILL.md Markdown 648L · 20.8 KB

安全亮点

✓ Security audit script is defensive in nature - checks credential permissions and scans for exposed secrets
✓ SKILL.md contains thorough security hardening guidance (prompt injection defense, external content warnings)
✓ Skill follows 'verify before done' pattern which reduces accidental exposure
✓ Clear documentation that external content is DATA not instructions
✓ No base64 encoding, eval(), or obfuscation techniques found
✓ No credential harvesting or data exfiltration code present
✓ Proper .gitignore guidance in security audit script