机器人程序用55,000个虚假注册淹没了我的反机器人初创公司。
Bots flooded my anti-bot startup with 55,000 fake signups

原始链接: https://humaverify.com/blog/anatomy-of-a-55k-bot-attack

一家名为 useHUMA 的反机器人初创公司最近遭受了大规模垃圾邮件攻击,其注册表单被 55,000 个虚假注册信息淹没。这件事充满了双重讽刺:该公司的主业正是销售机器人检测服务,且起初团队在查看数据时,还误以为这些垃圾流量是令人瞩目的业务增长。 事后调查揭示了三个主要的漏洞: 1. **缺乏频率限制(Rate Limiting):** 他们过度依赖蜜罐技术,而那些绕过用户界面的机器人程序轻易就避开了这些陷阱。 2. **“试吃”不当(Improper Dogfooding):** 他们虽然使用了自家的检测 API,但并未传输行为数据(鼠标/键盘输入模式),导致其安全防护形同虚设。 3. **电子邮件漏洞:** 每一个虚假注册都会触发一封自动邮件。由于邮件服务商的每日发送上限,该公司才侥幸避免了域名被列入黑名单。 该团队连夜修复了这些问题,通过实施严格的单 IP 频率限制、优化邮件逻辑,并最终在自己的注册流程中整合了完整的行为分析信号。 创始人总结道,伪装成“市场吸引力”的垃圾流量是一种危险的虚荣指标。通过公开记录这次失败,useHUMA 强化了一个关键教训:安全问题不能留到“以后”再解决,且产品必须完全按照既定方式使用才能发挥效用。

近期一篇 Hacker News 帖子讨论了一家名为 HumaVerify 的反机器人初创公司,该公司遭遇了 5.5 万次虚假注册。这些注册与土耳其赌场垃圾信息有关,通过触发海量验证邮件威胁到了该公司的基础设施,并导致其发送域名面临被列入黑名单的风险。 平台上的评论者对该帖子持怀疑态度,有人质疑这篇“诚恳”的文章本身是否由人工智能生成,或者仅仅是为了推销公司服务而精心策划的公关噱头。另一些人指出,这一事件凸显了“注册垃圾信息”的持续威胁,即恶意行为者利用合法服务来分发信息。一位用户指出,攻击者很可能使用了从数据泄露中抓取的真实电子邮件地址,以最大限度地扩大垃圾信息的传播范围,而非使用随机生成的地址。该讨论帖还触及了“回退诈骗(backscatter scams)”这一更广泛的问题,即垃圾邮件发送者利用支持工单系统来绕过电子邮件过滤器。
相关文章

原文
Anatomy of a 55,000-bot attack on an anti-bot startup | useHUMA

Post-mortem · June 11, 2026 · 7 min read

We sell bot detection. Last night I opened our database and found 555 pages of fake signups. This is the honest write-up: what broke, why it's embarrassing, how we fixed it in one night — and the accidental circuit breaker that saved our email domain.

The discovery

Tuesday, 11pm. I open the Supabase table editor to check something unrelated and the developers table feels… heavy. Every row has the same name: "🎁 Dene Hemen! 5K Lira Bonusunu Yakala" — Turkish for "Try it now! Grab the 5,000 Lira bonus." Casino spam. I scroll. Page 2. Page 40. Page 555.

58,618

junk verification logs

Yes — you read the third number right. We're early. For a few minutes our growth chart looked incredible, as long as you didn't read the names. The irony was not lost on me: bots flooded the signup form of a bot-detection product.

Root cause #1 — no rate limiting on signup

Our public API endpoints had per-plan rate limiting from day one. Our marketing site signup? A honeypot field and good intentions. Honeypots only catch bots polite enough to render your form — these ones POSTed the endpoint directly, thousands of times, and never even saw the hidden field.

A honeypot is a screen door. Rate limiting is the lock. We had the screen door.

Root cause #2 — we dogfooded wrong (the embarrassing one)

Here's the part that stings. Our signup did call our own bot-detection API on every registration. So why didn't it catch 55,000 bots?

Because we called it with just a userIdno behavioral signals. useHUMA scores behavior: mouse movement, keystroke cadence, scroll rhythm, touch patterns. Given zero signals, the engine returned a neutral score. Neutral cleared our threshold. Every bot passed.

We installed our own security camera and pointed it at the wall. If you dogfood your product, dogfood it the way you tell customers to use it — or it's decoration.

Root cause #3 — every fake signup sent a real email

Each registration fired a verification email. 55K signups = 55K attempted sends to fake addresses — the kind of bounce storm that gets a sending domain blacklisted. What saved us? Our email provider's free-tier daily cap. The limit we'd been mildly annoyed by quietly refused to send the flood. Domain reputation: intact.

Sometimes the constraint you resent is the circuit breaker you never installed.

The fix (same night)

Four changes, shipped before going to bed:

1. Per-IP rate limit on signup (5/min) — reusing the same
   atomic counter our API already used. ~12 lines.
2. Lifecycle emails (drip, trial reminders) now filter
   email_verified = true. Spam rows never get emailed again.
3. Deleted 113K junk rows (after pattern-matching them safely).
4. A weekly habit: one SQL count query. Monitoring you don't
   look at is monitoring you don't have.

Update, same week: shipped. Our signup now collects full behavioral signals client-side and scores them on every registration — and a submission carrying no signals at all gets rejected outright. Our own form finally uses useHUMA the way we tell customers to. The camera points at the door.

What I'd tell you if you run a signup form

  1. Rate limiting is layer zero. Not a hardening task for later — it's the lock on the front door. Per-IP, low ceiling, on every public POST.
  2. Honeypots don't stop direct-POST bots. They filter the lazy ones. Plan for the ones that never load your HTML.
  3. Dogfood properly or not at all. An anti-bot API called without behavior data is a coin flip wearing a lab coat.
  4. Your free-tier caps are accidental circuit breakers. Know which limits are protecting you before you upgrade them away.
  5. Spam looks like traction until you read it. 55,388 "signups" and 12 customers. Vanity metrics can literally be written by robots.
  6. Attacks are content. This post exists because of one. The most useful thing about a fire, once it's out, is the story of how it started.

The uncomfortable honesty, on purpose

We could have quietly fixed this. But useHUMA's whole pitch is that bot detection should be transparent — no black boxes, no PII, no theater. That has to include being transparent when the cobbler's children go barefoot. We got attacked, the attack found real gaps, the gaps are closed, and the product is better for it.

We verify humans by how they behave.

No CAPTCHAs, no puzzles, zero PII. One API call — now properly rate-limited, ask us how we know.

Try the live demoGet an API key

Written by Jaime Trejo, founder of useHUMA · Built in Mexico 🇲🇽

联系我们 contact @ memedata.com