中文 EN

扫描报告

扫描新文件

✓ 可信 — 风险评分 5/100
上次扫描：2 天前 重新扫描
5 /100
humanizer
Detect and remove AI writing patterns in text. Uses 24 pattern detectors, 500+ vocabulary terms, and statistical analysis (burstiness, type-token ratio, readability) to score text and suggest humanization fixes.
Pure text-processing Node.js skill with no runtime dependencies, no network access, no shell execution beyond invoking the CLI, and no sensitive filesystem access.
技能名称humanizer
分析耗时37.8s
引擎pi
✓
可以安装
This skill is safe to use. No security concerns identified.
资源类型声明权限推断权限状态证据
文件系统 READ READ ✓ 一致 cli.js reads input files via fs.readFileSync; declared in SKILL.md
网络访问 NONE NONE — No network imports (fetch/http) in any source file
命令执行 NONE NONE — scripts/*.sh only invoke 'node'; no subprocess or raw shell commands
环境变量 NONE NONE — No os.environ iteration or credential access
剪贴板 NONE NONE — No clipboard access
数据库 NONE NONE — No database access
技能调用 NONE NONE — No skill invocation chains
浏览器 NONE NONE — No browser automation
6 项发现
🔗
中危 外部 URL 外部 URL
https://img.shields.io/badge/License-MIT-yellow.svg
README.md:3
🔗
中危 外部 URL 外部 URL
https://img.shields.io/badge/tests-128%20passing-brightgreen
README.md:4
🔗
中危 外部 URL 外部 URL
https://img.shields.io/badge/node-%3E%3D18-brightgreen
README.md:5
🔗
中危 外部 URL 外部 URL
https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing
README.md:11
🔗
中危 外部 URL 外部 URL
https://arxiv.org/abs/2503.01659
README.md:11
🔗
中危 外部 URL 外部 URL
http://test.org/path?query=1&foo=bar
tests/edge-cases.test.js:163

目录结构

30 文件 · 206.6 KB · 5931 行
JavaScript 14f · 4486L Markdown 9f · 1336L JSON 2f · 54L Text 3f · 31L Shell 2f · 24L
├─ ▾ 📁 assets
│ └─ 📝 banner.md Markdown 10L · 331 B
├─ ▾ 📁 docs
│ ├─ 📝 CONTRIBUTING.md Markdown 69L · 2.0 KB
│ ├─ 📝 EXAMPLES.md Markdown 79L · 5.3 KB
│ └─ 📝 PATTERNS.md Markdown 59L · 2.4 KB
├─ ▾ 📁 references
│ ├─ 📝 ai-vocabulary.md Markdown 171L · 5.5 KB
│ ├─ 📝 patterns.md Markdown 332L · 10.0 KB
│ └─ 📝 style-guide.md Markdown 106L · 3.9 KB
├─ ▾ 📁 scripts
│ ├─ 🔧 analyze.sh Shell 12L · 301 B
│ └─ 🔧 humanize.sh Shell 12L · 315 B
├─ ▾ 📁 src
│ ├─ 📜 analyzer.js JavaScript 419L · 14.8 KB
│ ├─ 📜 cli.js JavaScript 574L · 19.6 KB
│ ├─ 📜 humanizer.js JavaScript 411L · 14.2 KB
│ ├─ 📜 patterns.js JavaScript 986L · 28.1 KB
│ ├─ 📜 stats.js JavaScript 275L · 9.5 KB
│ └─ 📜 vocabulary.js JavaScript 617L · 15.1 KB
├─ ▾ 📁 tests
│ ├─ ▾ 📁 fixtures
│ │ ├─ 📄 ai-sample-1.txt Text 15L · 2.2 KB
│ │ ├─ 📄 ai-sample-2.txt Text 7L · 1.1 KB
│ │ └─ 📄 human-sample-1.txt Text 9L · 1.1 KB
│ ├─ 📜 analyzer.test.js JavaScript 337L · 13.2 KB
│ ├─ 📜 calibration.test.js JavaScript 141L · 7.2 KB
│ ├─ 📜 edge-cases.test.js JavaScript 201L · 7.1 KB
│ ├─ 📜 humanizer.test.js JavaScript 174L · 6.4 KB
│ ├─ 📜 performance.test.js JavaScript 86L · 3.2 KB
│ └─ 📜 statistics.test.js JavaScript 226L · 7.9 KB
├─ 📋 _meta.json JSON 5L · 131 B
├─ 📜 eslint.config.js JavaScript 31L · 559 B
├─ 📋 package.json JSON 49L · 1.3 KB
├─ 📝 README.md Markdown 361L · 15.8 KB
├─ 📝 SKILL.md Markdown 149L · 7.8 KB
└─ 📜 vitest.config.js JavaScript 8L · 152 B

依赖分析 3 项

包名版本来源已知漏洞备注
eslint ^9.0.0 npm dev 否 Dev dependency only; not bundled with skill
prettier ^3.0.0 npm dev 否 Dev dependency only; not bundled with skill
vitest ^3.0.0 npm dev 否 Dev dependency only; not bundled with skill

安全亮点

✓ No runtime external dependencies — only devDependencies for linting and testing
✓ No network requests in any source file
✓ No subprocess, exec, or shell command execution beyond invoking node
✓ No credential, SSH, AWS, or .env file access
✓ No eval, Function(), or dynamic code execution
✓ No base64 encoding/decoding with shell piping
✓ No hidden instructions or obfuscated code
✓ Doc-to-code alignment is strong — declared capabilities match implementation
✓ No data exfiltration channels
✓ Shell scripts are simple wrappers with no curl/wget remote execution