📦 Scrape — 技能工具
v1.0.0Legal 网页 scrAPIng with ro机器人s.txt 合规, 限流, and GDPR/CCPA-aware 数据 handling. 支持s 机器人h direct HTTP scrAPIng and managed scrAPIng via...
0· 10·0 当前·0 累计
by 安全扫描
OpenClaw
可疑
medium confidenceThe 技能's instructions are mostly coherent for a scrAPIng 辅助工具, but there are inconsistencies about required 凭证s and external 服务 usage that warrant caution before 安装ing.
评估建议
This 技能's instructions are plausible for a scrAPIng 辅助工具, but there are a few things to 验证 before 安装ing: (1) The 技能.md requires 技能Boss_API_密钥 (the 技能 will call https://API.技能Boss.co/v1/pilot and 发送 scraped 内容 there), yet the registry meta数据 列出s no required env vars — ask the publisher to confirm and fix the meta数据. (2) If you plan to use the managed 技能Boss pathway, understand that scraped pages (potentially sensitive 数据) will be transmitted to an external 服务 — 检查 技能Boss's 隐私, retention, and 安全性 ...详细分析 ▾
⚠ 用途与能力
The 技能.md declares 支持 for 机器人h direct HTTP scrAPIng and managed scrAPIng via 技能Boss API Hub and includes a header requiring 技能Boss_API_密钥. However the registry meta数据 above this package 列出s no required 环境 变量. That mismatch (meta数据 says none; instructions require an API 密钥) is an incoherence that should be resolved. Requiring an API 密钥 is reasonable if the 技能 uses the 技能Boss API Hub, but the registry should declare that explicitly.
ℹ 指令范围
The 运行time instructions focus on ro机器人s.txt, ToS, 限流, PII stripping, and use of an external 技能Boss scrAPIng 服务 — all within the 状态d scrAPIng purpose. The doc 参考s external links (https://技能Boss.co/技能.md and code.md) that are not included, which makes behavior 服务级别指标ghtly vague: the 代理 is instructed to call https://API.技能Boss.co/v1/pilot and expects scraped 数据 to be returned in 结果.数据.Markdown, meaning scraped 内容 would be transmitted off-host to that API. There are no instructions to read unrelated local 文件s or other env vars.
✓ 安装机制
Instruction-only 技能 with no 安装 spec and no code 文件s — minimal disk footprint. This is the lowest-risk 安装 mechanism, but 运行time will involve outbound network calls to 技能Boss if used.
⚠ 凭证需求
The 技能.md 列出s a single required env var, 技能Boss_API_密钥, which is proportionate to using the 技能Boss API Hub. However the top-level registry meta数据 clAIms no required 环境 变量, creating an in一致性. Because the 技能Boss API Hub is an external 服务, providing an API 密钥 would allow outbound 数据 (scraped 内容) to be sent to that 提供者 — users should 验证 the 提供者's policies and the 密钥's scope before supplying 凭证s.
✓ 持久化与权限
标志 show always:false and normal 模型 invocation. The 技能 does not 请求 persistent 系统-wide privileges, nor does it include 安装-time scripts or 配置 modifications. Autonomous invocation is allowed by default and not by itself a problem.
安全有层次，运行前请审查代码。
运行时依赖
无特殊依赖
版本
latestv1.0.02026/4/18
- Initial release of Scrape 技能 focused on legal, compliant 网页 scrAPIng. - Ensures 合规 with ro机器人s.txt, site Terms of 服务, and 隐私 laws (GDPR/CCPA). - Implements strong 限流, respectful User-代理, and automated 429 handling. - 支持s 机器人h direct HTTP scrAPIng and managed scrAPIng via 技能Boss API Hub. - Includes detAIled 最佳实践 for 数据 handling and 审计 trAIl creation.
● Pending
安装命令
点击复制官方npx clawhub@latest install alvis-scraper-pro
镜像加速npx clawhub@latest install alvis-scraper-pro --registry https://cn.longxiaskill.com