Show HN: Replacing CAPTCHAs with adaptive proof-of-work in WordPress
来源:Hacker News 首页收录:2026/10/22
✦ AI 解读
作者为企业开发举报服务时,为符合欧盟隐私法规并低成本拦截垃圾机器人,采用 Hashcash 工作量证明机制,实现无 Cookie、无追踪、轻量且对访客不可见的反垃圾方案,实测拦截率 100% 且未误伤真实提交。因生态中缺乏免费全功能方案,作者将其封装为免费 WordPress 插件开源发布。
When I founded my business a whistle-blowing-service for enterprises, I’ve seeked for a working solution to block spam bots on that home page. The solution was intended to comply with EU data privacy law, no cookies, no sniffing, light-weight-size, on my webserver only, beeing invisible for my site visitors and no running costs at all.I came across Hashcash, a proof-of-work approach, to block spam. When I implemented it, it worked immediately and blocked 100 percent without blocking any incoming real posts.I was astonished, that no solution existed in the ecosystem allready that covered all forms and was for free. Therefore I‘ve turned it into a free wordpress-plugin.It is not only a technical solution, but starts with the business model of spam (mass spamming at low cost), by producing a calculation effort for every single post, that unavoidable. And therefore is not exactly unbreakable but becomes very expensive for the spammer and therefore alows only for targeted high quality approaches that need other treatment. Up to now my own page remains spam free and users report the same.Now, running since 2022 on 4000+ active installations, it is working with almost every webform and plugin combination that is running on any wordpress instance, using pattern recognition on http-post-level instead of specific proprietary integration for single specific form builders. It ships with 80+ starting patterns for the most relevant form builders and themes, but can be enhanced with any pattern by the user.The plugin can be previewed directly and immediately on wordpress.org. It ships with14 KB compressed under a GPL-2.0 license.I’m looking forward to discuss about the proof-of-work approach, form detection, privacy implications, and tradeoffs.
Comments URL: https://news.ycombinator.com/item?id=50194762
Points: 5
# Comments: 0
相关推荐
H2oai/jdupes: H2O.ai fork of jdupes
✦ AI 摘要H2O.ai 在 GitHub 上发布了 jdupes 的分支版本。jdupes 是一个用于查找和去重重复文件的命令行工具,H2O.ai 的分支可能包含针对其内部需求的修改或增强。该项目在 Hacker News 上仅获得 1 分且无评论,关注度较低,但对需要文件去重功能的开发者可能有一定参考价值。
Open Source Mobile-Use MCP and AI Agent
✦ AI 摘要这是一个在 Hacker News 上分享的开源项目 buddy-android,提供移动端使用的 MCP 和 AI Agent 实现。项目托管于 GitHub,目前关注度较低(1 分、1 条评论),但为在 Android 设备上构建和运行 AI 智能体提供了开源参考,适合对移动端 Agent 和 MCP 协议感兴趣的开发者研究。
Show HN: Koala – Run AI Agents in Isolated Linux VMs on macOS
✦ AI 摘要Koala 是一个在 Hacker News 上展示的开源项目,旨在让用户在 macOS 上通过隔离的 Linux 虚拟机运行 AI 智能体。它提供了安全沙箱环境,避免智能体直接访问宿主机系统,适合需要隔离执行 AI 代理任务的开发者。项目通过 Homebrew 分发,目前关注度较低。