惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

Google DeepMind News
Google DeepMind News
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
酷 壳 – CoolShell
酷 壳 – CoolShell
WordPress大学
WordPress大学
小众软件
小众软件
博客园 - 司徒正美
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
Jina AI
Jina AI
Hugging Face - Blog
Hugging Face - Blog
博客园 - Franky
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
量子位
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
雷峰网
雷峰网
云风的 BLOG
云风的 BLOG
M
MIT News - Artificial intelligence
F
Fortinet All Blogs
T
Tailwind CSS Blog
Martin Fowler
Martin Fowler
I
InfoQ
The GitHub Blog
The GitHub Blog
有赞技术团队
有赞技术团队
The Cloudflare Blog
罗磊的独立博客

DEV Community

Authentication Security Deep Dive: From Brute Force to Salted Hashing (With Java Examples) Why AI Systems Don’t Fail — They Drift Spilling beans for how i learn for exam😁"Reinforcement Learning Cheat Sheet" I Replaced Chrome with Safari for AI Browser Automation. Here's What Broke (and What Finally Worked) How Python Borrows Other People's Work The $40 Architecture: Processing 1 Billion API Requests with 99.99% Uptime Vibe Coding: A Workflow Guide (From Zero to SaaS) Most webhook security guides protect the wrong side. The scary part is delivery. Headless CMS for TanStack Start: Build a Blog with Cosmic EU Age Verification App "Hacked in 2 Minutes" — What Actually Happened Comfy Cloud’s delete function does not actually remove files Running AI Models on GPU Cloud Servers: A Beginner Guide Event-driven media intelligence with AWS Step Functions and Bedrock I scored 500 AI prompts across 8 quality dimensions — here's what broke How to Call Google Gemini API from Next.js (Free Tier, No Backend Needed) The Portal Protocol: Reclaiming Human Connection in the Age of AI How to Fix Your Team's Scattered Knowledge Problem With a Self-Hosted Forum Intro to tc Cloud Functors: A Graph-First Mental Model for the Modern Cloud Designing Multi-Tenant Backends With Both Ownership and Team Access I Built a Neumorphic CSS Library with 77+ Components — Here's What I Learned PostgreSQL Performance Optimization: Why Connection Pooling Is Critical at Scale Cómo construí un SaaS multi-rubro para gestionar expensas en Argentina con FastAPI + Vue 3 🚀 I Built an Ethical Hacking Scanner Tool – Open Source Project I Replaced /usage and /context in Claude Code With a Single Statusline A Pythonic Way to Handle Emails (IMAP/SMTP) with Auto-Discovery and AI-Ready Design I Collected 8.9 Million Polymarket Price Points — Here's What I Found About How Markets Really Move EcoTrack AI — Carbon Footprint Tracker & Dashboard Everyone's Using AI. No One Agrees How. 5 self-hosted ebook managers worth trying in 2026 Building Your First AI Agent with LangChain: From Chatbot to Autonomous Assistant
The Kimwolf Bust Just Outed Android Webcams as Botnet Fod...
Super Funicular · 2026-05-25 · via DEV Community

Super Funicular

On May 22, 2026, the U.S. Department of Justice announced the arrest of a 23-year-old in Ottawa for allegedly running Kimwolf, a DDoS-for-hire botnet that prosecutors say issued more than 25,000 attack commands and helped power record-setting floods peaking at 31.4 terabits per second. According to the indictment, Kimwolf is a variant of AISURU that specifically targeted Android devices with an exposed Android Debug Bridge (ADB) service — and the device class the DoJ called out by name was striking: "digital photo frames and web cameras."

That language matters. The Kimwolf victims weren't enterprise servers or IoT toys built by sketchy vendors no one has heard of. They were Android-powered cameras and frames — the same form factor millions of people are now building on purpose, by reaching for the old phone in a drawer and turning it into a security camera, a nursery monitor, or a wildlife cam. The Hacker News covered the arrest here; the DoJ press release is here. Forty-five DDoS-for-hire storefronts were seized in parallel.

If you've repurposed an Android phone as a camera — or you're thinking about it — this is a moment to look at your setup and ask the question the DoJ is essentially forcing on the camera-app industry: what services on this phone are reachable from outside your network, and who is on the other end of them?

Why this isn't a "patch your firmware" story

The reflexive read of any IoT botnet news is: "vendors should patch faster." Kimwolf is different. There is no obscure camera firmware here. The exploitation surface is Android Debug Bridge — a feature that ships with every Android phone, sits dormant by default, and only becomes dangerous when (a) it's turned on, and (b) the device is on a network that lets the outside world reach it.

That means Kimwolf isn't a story about a single buggy product. It's a story about an entire category of consumer devices — Android phones being used as cameras — where the architecture of the camera app and the network around it determines whether the device is a useful tool or a unit in someone's 31-Tbps DDoS army.

Three things have to be true for a phone-camera setup to end up in that army:

  1. ADB is enabled (often because the user followed an old tutorial that said "enable developer mode to keep the screen off while streaming").
  2. The phone is reachable from the open internet (either via UPnP-mapped ports, an exposed router, or because the user wired it up for "remote viewing" without thinking about who else can reach it).
  3. The camera app needs either of the above to function.

It's that third point that separates a defensible setup from a recruitable one.

The architecture question every repurposed-phone camera setup now has to answer

Strip away the indictment language and Kimwolf is a stress test of three choices any phone-camera product can be measured against:

1. Does the app require ADB to be enabled?
A camera app that needs Developer Options on and ADB authorized in order to keep recording with the screen off is asking the user to widen the attack surface. The opposite end of the spectrum is an app that uses Android's documented Camera2 + foreground-service APIs to record with the screen off — no developer mode, no ADB, no extra daemons.

2. Does the app push video to the cloud by default?
Cloud-by-default architectures push the user toward an account, a remote viewing endpoint, and an inbound or relayed connection. Local-by-default architectures keep the footage on the device unless and until the user explicitly does something to share it.

3. Does the app require an open inbound port to be useful?
"Remote viewing" features that work by mapping a port through the router so a remote client can hit the phone directly are the textbook pattern that gets a device picked up by network-wide scans like the ones AISURU/Kimwolf operators run.

A camera setup that answers "no" to all three is structurally hard to recruit. A setup that answers "yes" to any one of them is on the menu.

What this means for the "old phone as a camera" trend

There's a healthy and growing community of people repurposing old Android phones as cameras — for nurseries, garages, workshops, wildlife observation, lecture recording, dashcams, you name it. The reuse story is great: keep a working device out of the landfill, get a camera at zero hardware cost, control your own footage.

The Kimwolf bust is the moment to put a finer point on the how. The phones-as-cameras movement only stays good if the underlying app respects three constraints:

  • No "enable Developer Options and ADB to make it work" workarounds. That step alone is what put thousands of Android devices into Kimwolf.
  • Local storage as the default. Footage should sit on the phone, not on a vendor server, unless the user opts in to streaming.
  • No required inbound ports. If "remote viewing" is offered, it should be initiated by the user from the viewing device — not by punching a hole in the firewall the camera is sitting behind.

This is exactly the architecture we built Background Camera RemoteStream around. It records with the screen off using documented Android APIs — no ADB, no developer mode, no daemon you have to keep alive. Footage is stored locally on the phone by default; nothing is uploaded to a third-party server. Streaming is opt-in and routed through YouTube Live's own infrastructure, which means no inbound port on your home network, no account-system honeypot, and no plaintext RTSP endpoint sitting on the open internet waiting for the next scanner sweep.

That's the structural posture the Kimwolf indictment makes a value proposition. If your camera app needs you to widen your attack surface, the Kimwolf operators were literally selling access to the result.

What to do today, even if you don't change apps

Three concrete checks for anyone running a repurposed Android camera right now:

  1. Open Settings → System → Developer Options. If ADB Debugging is on and you don't actively use adb, turn it off.
  2. Check your router. Look at port forwarding and UPnP-mapped ports. If you see a forwarded port pointing at the phone's local IP that you don't recognize, drop it.
  3. Audit your camera app's permissions and network behavior. If the app needs developer mode to function, that's the signal to look for an app that doesn't.

Where to go from here

Kimwolf is one indictment in one country. The takedown also dismantled command-and-control infrastructure for AISURU, JackSkid, and Mossad — meaning the pattern is broader than one operator, and the next variant will look at the same Android-camera surface again.

If you want to repurpose an old phone as a camera without renting a slot in someone's DDoS botnet, the architecture matters more than the brand. Pick an app that records locally, doesn't need ADB, and doesn't require an open inbound port. That's the brief.

Background Camera RemoteStream is on Google Play: https://play.google.com/store/apps/details?id=com.superfunicular.digicam

More on our approach: https://superfunicular.com

Related reading


Sources: