扫描报告
15 /100
deepread-legal
Extract structured data from contracts, legal agreements, court filings, and compliance documents. Pre-built schemas for parties, clauses, dates, obligations. PII redaction for privilege review.
A legitimate legal document extraction skill that sends documents to a declared third-party API. No malicious behavior detected; API key placeholders in docs are benign examples.
可以安装
Approve for use. Users should be aware that sensitive legal documents are transmitted to an external service (api.deepread.tech) — verify compliance requirements before processing confidential materials.
安全发现 2 项
| 严重性 | 安全发现 | 位置 |
|---|---|---|
| 低危 | API key placeholder examples in documentation | SKILL.md:55 |
| 提示 | External document processing | SKILL.md:14 |
| 资源类型 | 声明权限 | 推断权限 | 状态 | 证据 |
|---|---|---|---|---|
| 网络访问 | READ | READ | ✓ 一致 | SKILL.md:14 - POSTs documents to api.deepread.tech |
| 环境变量 | READ | READ | ✓ 一致 | SKILL.md:metadata - requires DEEPREAD_API_KEY env var |
| 文件系统 | NONE | NONE | — | No file operations beyond reading uploaded documents |
| 命令执行 | NONE | NONE | — | No shell execution detected |
2 高危 10 项发现
高危 API 密钥 疑似硬编码凭证
API_KEY="sk_live_your_key_here" SKILL.md:55 高危 API 密钥 疑似硬编码凭证
API_KEY = "sk_live_YOUR_KEY" SKILL.md:123 中危 外部 URL 外部 URL
https://www.deepread.tech SKILL.md:5 中危 外部 URL 外部 URL
https://api.deepread.tech SKILL.md:12 中危 外部 URL 外部 URL
https://www.deepread.tech/dashboard/?utm_source=clawhub SKILL.md:50 中危 外部 URL 外部 URL
https://api.deepread.tech/v1/process SKILL.md:201 中危 外部 URL 外部 URL
https://www.deepread.tech/dashboard/byok SKILL.md:273 中危 外部 URL 外部 URL
https://www.deepread.tech/dashboard SKILL.md:287 中危 外部 URL 外部 URL
https://www.npmjs.com/package/n8n-nodes-deepread SKILL.md:289 提示 邮箱 邮箱地址
[email protected] SKILL.md:291 目录结构
1 文件 · 12.6 KB · 295 行 Markdown 1f · 295L
└─
SKILL.md
Markdown
安全亮点
✓ Behavior is fully declared in SKILL.md - no hidden functionality
✓ No shell execution, subprocess, or system command invocation
✓ No credential harvesting beyond the required API key
✓ No base64 encoding, eval patterns, or obfuscated code
✓ No access to sensitive local paths (~/.ssh, ~/.aws, .env)
✓ No suspicious network activity beyond the declared API endpoint
✓ Uses standard HTTPS API calls with proper authentication headers