CPU mining with XMRig on Windows, Linux and macOS: pool or solo.

CPU mining with XMRig on Windows, Linux and macOS: pool or solo.

CPU mining with XMRig on Windows, Linux and macOS: pool or solo.

submitted by dMartian to monero 3 points | 0 comments https://github.com/d-Martian/Kind-Miner Hi everyone. I made a monero mining app that I think you will be interested in. It is based on cypherpunk ethos, using xmrig/p2pool, and is FOSS w/ reproducible builds. It’s a mining app unlike any other because of how it’s designed to be used by casual users and pros alike with minimal fuss. Under the hood it works much like Gupax, but it is much simpler for the users and less maintenance. It has unique features to “stay out of your way” on your daily driver, making optimizations for user experience over mining performance. It still does a very good job! I made it so that users can easily can choose how much they want to strain their system There’s a lot I can say about the app, but AI did a pretty good job of doing that for me: I’ve been building **Kind Miner** , a Monero miner for the computers you already leave on. It runs at low priority, backs off the moment you need your machine, and pays straight to your wallet through P2Pool. No pool account, no fees, no telemetry. github.com/d-Martian/Kind-Miner (MIT) Wiring up P2Pool, XMRig, a node, and Tor is a weekend project, and then you still have to watch it so it doesn’t cook your laptop or lag your game. Kind Miner handles the plumbing and the babysitting behind a four-step setup and a tray icon. Someone who has never opened a config file can mine on P2Pool over Tor, and someone who reads source can verify every byte. **Under the hood** * **XMRig built from source** with two patches, so there’s no built-in 1% donation and there’s a native linux-arm64 build * **P2Pool** for payouts: you’re paid in the coinbase output, so there’s no operator, no payout threshold, no account * **Tor by default** between you and the node (bundled on desktop builds). Defaults to a dedicated Nodo; point it at your own monerod or any remote node if you prefer * **Throttling that doesn’t fight RandomX:** every 2 seconds it measures non-miner CPU load and pauses/resumes XMRig with SIGSTOP/SIGCONT instead of changing thread counts, so you never pay the dataset re-init cost * **Hybrid-CPU aware:** efficiency cores while you’re at the keyboard, performance cores when you walk away * **On Linux:** idle CPU scheduling, idle I/O, and a high OOM score so the miner is the first thing killed under pressure * **One native Go binary** (Fyne GUI). No Electron, no runtime **Reproducible builds** You’re trusting this with your wallet address and your network traffic, so you shouldn’t have to trust my build machine. The Go toolchain and dependencies are pinned, nondeterminism is stripped out, and Linux GUI/.deb builds run in a pinned, offline Debian 12 container. XMRig gets the same treatment (pinned sources, pinned Alpine toolchain, no network). `scripts/rebuild-release.sh <tag>` rebuilds a release and checks it against the published hashes. macOS, Windows, AppImage, and Flatpak aren’t reproducible yet; details in REPRODUCIBLE.md. **Kind to the machine it’s on** * Four presets: Ghost (52%), Polite (78%, default), Balanced (90%), Full * Stays at the lowest ceiling until you’ve been idle 5 minutes; “Mine now” skips the wait * Pauses on battery, on high temps or thermal throttling, and while Feral GameMode is running * Stops completely under memory pressure, restarts after 10 calm minutes * Snooze for an hour, until 06:00, or until you resume * Warns you if your P2Pool node is on an island chain or Wi-Fi power-save is adding latency **Runs on almost anything you leave on** * **Desktops and laptops:** GUI with setup wizard, live resource chart, and a color-coded tray icon. Linux is the main target; macOS and Windows work but are less tested * **Packages:** AppImage, Flatpak, .deb, or a bare binary that fetches SHA256-pinned engines * **Home servers and Nodos:** `kind-minerd` is a static headless binary for x86-64 and arm64, as a systemd service or a .deb that self-updates over Tor * **Terminal people:** text wizard and a terminal mode with stdout logs * Needs about 2 GB of RAM for the RandomX dataset and a Monero address **Privacy notes** No telemetry, analytics, update pings, accounts, or registration. Pool mode was removed entirely because it exposed wallet addresses and paid fees. As with any P2Pool setup, your address is public on the sidechain, so use a dedicated mining wallet. **Looking for** * Testers on macOS, Windows, and arm64 * Independent rebuilders willing to co-sign release hashes * Bug reports, issues, and PRs Thanks for reading. Happy to answer anything in the comments.

Claude Code localiza un minero XMRig en un Mac tras analizar lecturas térmicas críticas. Claves del cryptojacking y la seguridad en macOS.

Check Point Research reports a two-month campaign (Yoba) using fake GitHub projects posing as AI desktop apps (GPT-5.6 Sol, GPT-6 Astra, Claude Opus), placing malware beside legitimate signed OpenAI and Anthropic executables for DLL sideloading; the observed final payload was an XMRig cryptominer,…

🔍 Contexte : Le 29 octobre 2026, Pluto Research (Pluto Security) publie une analyse technique détaillée d'une campagne malveillante exploitant les dépôts de modèles IA sur Hugging Face et ModelScope. L'investigation a été menée par Shaked Chen, Eitan Chassid et Yotam Perkal.

Check Point describes Yoba, a campaign using fake ChatGPT and Claude desktop-app projects to distribute YobaLoader and XMRig miners, with DLL sideloading…

AI Tool Impersonation Becomes the Entry Point for a Sophisticated Malware Campaign Cybercriminals are increasingly exploiting the popularity of artificial intelligence to make malware distribution look like legitimate software installation. A newly documented campaign demonstrates how easily trusted developer platforms, recognizable AI product names, and search engine rankings can be combined to trick unsuspecting users into installing malicious applications. Check Point Research has spent two months investigating a campaign dubbed Yoba, which distributes a custom malware chain through fraudulent GitHub projects impersonating popular AI desktop applications.

At a Glance Malware Family 7 Android RAT families (cyb3rx, appkjl, godz, d-passive, d-shine, GhostBat/GhostGrab, and tenxen) paired with the d-miner XMRig dropper Threat Actor Confirmed analytical cluster tracked as CyberDakait; suspected multiple independent sub-project developers (no formal named group attributed) Target / Victims

Fake ChatGPT and Claude desktop apps on GitHub use DLL sideloading to deploy YobaLoader, which retrieves and executes an XMRig miner.

Fake ChatGPT and Claude desktop apps on GitHub use DLL sideloading to deploy YobaLoader, which retrieves and executes an XMRig miner.

Check Point Research details Yoba, a campaign using fake ChatGPT and Claude GitHub apps and signed binaries for DLL sideloading. Its loader uses…

At a Glance Malware Family 7 Android RAT families (cyb3rx, appkjl, godz, d-passive, d-shine, GhostBat/GhostGrab, and tenxen) paired with the d-miner XMRig dropper Threat Actor Confirmed analytical cluster tracked as CyberDakait; suspected multiple independent sub-project developers (no formal named group attributed) Target / Victims