github.com
Show HN: Auto-Co – 14 AI agents that run a startup autonomously (open source)https://github.com/NikitaDmitrieff/auto-co-metaAuto-Co is an autonomous AI company OS — not a framework you build on, but a running system with an opinionated structure.Architecture: - 14 agents with expert personas (CEO/Bezos, CTO/Vogels, CFO/Campbell, Critic/Munger...) - Bash loop + Claude Code CLI — no custom inference, no vector stores - Shared markdown consensus file as the cross-cycle relay baton - Human escalation via Telegram for true blockers only (2 escalations in 12 cycles) - Every cycle must produce artifacts: code, deployments, docsThe repo IS the live company. It built its own landing page, README, Docker stack, GitHub release, and community posts — all autonomously across 12 cycles of self-improvement.What makes it different from AutoGen/CrewAI/LangGraph: those are building blocks. Auto-Co is the building. The decision hierarchy, safety guardrails, and convergence rules are baked in. You give it a mission and a Claude API key; it runs.The Critic agent (Munger persona) has been the most valuable: it runs a pre-mortem Mar 6, 2026 9:48 PM
news.ycombinator.com
Launch HN: Palus Finance (YC W26): Better yields on idle cash for startups, SMBshttps://news.ycombinator.com/item?id=47278980Hi HN! We’re Sam and Michael from Palus Finance (https://palus.finance). We’re building a treasury management platform for startups and SMBs to earn higher yields with a high-yield bond portfolio.We were funded by YC for a consumer-focused product for higher-yield savings. But when we joined YC and got our funding, we realized we needed the product for our own startup’s cash reserves, and other startups in the batch started telling us they wanted this too.We realized that traditional startup treasury products do much the same thing: open a brokerage account, sweep your cash into a money market fund (MMF), and charge a management fee. No strategy involved. (There is actually one widely-advertised treasury product that differentiates on yield, but instead of an MMF it uses a mutual fund where your principal is at considerable risk – it had a 9% loss in 2022 that took years to recover.)I come from a finance background, so this norm felt weird to me. The typical startup cashflow pattern isMar 6, 2026 6:26 PM
github.com
Show HN: Code Jump Tracker – Never get lost after "Go to Definition" againhttps://github.com/tominaga-h/code-jump-trackerHi HN,I build a VS Code extension called Code Jump Tracker. https://github.com/tominaga-h/code-jump-trackerI created this because I was constantly getting lost in large codebases during code reviews. I use "Go to Definition" (F12) a lot to trace logic. While VS Code has a native "Navigate Back" shortcut, it operates blindly—you can't see your stack history, making it hard to jump back to a specific file after exploring multiple branches.Code Jump Tracker solves this by acting like `pushd/popd` for your editor.Key features: - Auto-tracking: Automatically logs your jumps (Go to Definition, Outline clicks) into a clean, native-looking Webview table in the sidebar. - Manual Pinning: Found a core function you need to revisit? Pin it. It adds a persistent marker icon to your gutter. - Unique Locations: Keeps a deduplicated list of your visited symbols. - Persistent: State is saved per-workspace using `workspaceState`, so your history survives editor restarts.I built the UI using VS Code's naMar 6, 2026 5:27 PM

spacenews.com
Eutelsat completes $5.8 billion refinancing planhttps://spacenews.com/eutelsat-completes-5-8-billion-refinancing-plan/Eutelsat has completed the last step in a 5 billion euro ($5.8 billion) refinancing plan to refresh its OneWeb constellation and support Europe’s IRIS² sovereign connectivity program, the French satellite operator announced March 6. The post Eutelsat completes $5.8 billion refinancing plan appeared first on SpaceNews.Mar 6, 2026 5:16 PM
github.com
Show HN: RapidFire AI – parallel RAG experimentation with live run interventionhttps://github.com/RapidFireAI/rapidfireaiWe built RapidFire AI because iterating on RAG pipelines is painfully sequential: run a config, wait, inspect results, tweak one knob, repeat. When you have 15 things to tune (chunk size, retrieval k, reranker, prompt template, context window strategy...) that cycle compounds fast.RapidFire uses shard-based interleaved scheduling to run many configurations concurrently on a single machine — even a CPU-only box if you're using a closed API like OpenAI. Instead of config A finishing before config B starts, all configs process data shards in rotation, so you see live side-by-side metric deltas within the first few minutes.The part we're most excited about: Interactive Control (IC Ops).Most RAG observability tools tell you what happened after a run finishes. IC Ops closes the loop — you can act on what you're observing mid-run: - Stop a config that's clearly underperforming (save the API spend) - Resume it later if you change your mind - Clone a promising run and modify its prompt templateMar 6, 2026 5:12 PM
qurt.ai
Show HN: Qurt – open-source desktop AI coworker (BYOK, agent mode)https://qurt.ai/Hi HN — I built Qurt, an open-source desktop “AI coworker” for Windows + macOS.The goal: provider freedom + local-first UX. You bring your own API keys (no subscription / no lock-in), and your chats/keys/files stay on your machine (nothing stored on our servers).What it does: - BYOK + multi-provider model switching (keep control of keys/data) - Desktop-native chat with file/image attachments - Agent mode: web browsing/search, terminal commands, browser automation, and file edits — with “review before run”Downloads (GitHub Releases): https://github.com/qurtai/qurt/releases/tag/v0.1.1Repo: https://github.com/qurtai/qurtLanding page: https://qurt.ai/I’d love feedback on: 1) the “review-before-run” UX (too noisy? not strict enough?) 2) what provider/model integrations you’d want next 3) what would make you trust/install this faster (docs, sandboxing details, etc.)Mar 6, 2026 4:53 PM
github.com
Show HN: Claude skill to do your taxeshttps://github.com/robbalian/claude-tax-filingTL;DR Claude Code did my 2024 and 2025 taxes. Added a skill that anyone can use to do their own.I tested it against TurboTax on my own 2024 and 2025 return. Same result without clicking through 45 minutes of wizard steps.Would love PRs or feedback as we come up on tax season.Learnings from replacing TurboTax with ClaudeSkill looping The first iteration of my taxes took almost an hour and a decent amount of prompting. Many context compactions, tons of PDF issues, lots of exploration. When it was done, I asked Claude to write the skill to make it faster the next time (eg: Always check for XFA tags first when discovering form fields)Skills aren’t just .MD files anymore Now they’re folders that can contain code snippets, example files, rules. In 2025 we were all building custom agents with custom tools. In 2026 every agent has its own code execution, network connection, and workspace. So it’s custom skills on the same agent (trending towards Claude Code or Cowork)Mar 6, 2026 4:32 PM
github.com
Show HN: Codaholiq, AI automations for GitHub repositorieshttps://github.com/Njuelle/CodaholiqHi HN,I kept finding myself writing scripts or manually running prompts to do things like PR reviews, documentation generation, or issue triage on my GitHub repositories.After doing this enough times, I decided to build a small platform to automate these workflows.I recently open sourced it:https://github.com/Njuelle/CodaholiqCodaholiq lets you run AI-powered workflows triggered by GitHub events.You connect a repository and define automations that run when certain events occur, such as:- pull request opened - push to a branch - issue created - cron schedule - manual triggerEach automation defines:- a trigger - an AI provider or model (Claude Code, Codex, Gemini, etc.) - a prompt template - optional conditions on the GitHub event payloadWhen the event happens, Codaholiq renders the prompt using variables from the event context and dispatches a GitHub Actions workflow that runs the AI task.The platform also tracks executions with:- real-time log streaming - token usage and cost tracking Mar 6, 2026 4:08 PM
github.com
Show HN: GTAO and Red Dead Online lag switch detector and blocker (anti-cheat)https://github.com/Sovandeulv/killswitchRed Dead Redemption 2 is a beautiful game, and its PvP used to be genuinely fun if you didn't take it too seriously. That has changed. Both RDO and GTA Online lobbies are full of cheaters, and on console, lag switching is the go-to method. A lag switcher briefly halts their own connection, moves and shoots while your hits get dropped, then reconnects. You land a clean headshot and it doesn't register. Their shots land fine. Die, respawn, repeat.Both games use P2P mesh networking, so every player connects directly to every other player. Your IP is exposed to the entire lobby, and anyone can manipulate their connection to gain a combat advantage. Rockstar has never added anti-cheat to console.Killswitch sits on a Mac bridging your router and console, monitoring packet timing on the primary game port (UDP 6672). All game traffic is encrypted, so detection relies entirely on metadata and timing gaps between packets. When an IP's gap score crosses a threshold, it gets blocked via the macOS Mar 6, 2026 4:00 PM

spacenews.com
Commercial Space Federation (CSF) Welcomes New Membershttps://spacenews.com/commercial-space-federation-csf-welcomes-new-members/March 6, 2026 – Washington, D.C.—The Commercial Space Federation (CSF) is pleased to welcome Leolabs, the American Society for Gravitational and Space Research (ASGSR), and SurgeStreams. Together, these organizations strengthen […] The post Commercial Space Federation (CSF) Welcomes New Members appeared first on SpaceNews.Mar 6, 2026 4:00 PM

nasa.gov
Weekends on the Space Stationhttps://www.nasa.gov/image-article/weekends-on-the-space-station/NASA astronaut Jessica Meir trims the hair of fellow NASA astronaut Jack Hathaway in this March 1, 2026, image. Meir uses an electric razor attached to a vacuum that collects loose clippings to keep the station’s atmosphere clean in microgravity. Crew on the International Space Station also use weekends to complete housekeeping tasks. Learn more […]Mar 6, 2026 3:23 PM
github.com
Show HN: Stateless Go GraphQL Federation Gateway, 1.58x Faster Than ApolloRouterhttps://github.com/n9te9/go-graphql-federation-gatewayI built a GraphQL Federation v2 gateway in pure Go.What's supportedAll core Federation v2 directives work out of the box:- @key — simple, composite, nested ("coord { lat lng }"), multiple @key per type, @key(resolvable: false) - @external — correct ownership filtering; stub references (e.g. a service that declares a type only to satisfy a key) are never mistaken for entity owners - @requires — chained dependency injection resolved in topological order - @provides — skips unnecessary _entities fetches when the parent service already covers all requested child fields - @shareable — prefers the parent subgraph to avoid extra entity fetches - @override — field ownership migration between subgraphs - @inaccessible — enforced at both schema composition and query validation - @interfaceObject — interface types as entities with inline fragment resolution - @composeDirective — custom directive preservation with consistency validation - @tag, @linkOperations: query and mutation. Independent subgMar 6, 2026 3:08 PM
bing.com
Dan Piazza Up For Title Defense At National Horseplayers Championshiphttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8cb75d61ed47c3b61273dec02dd38a&url=https%3A%2F%2Fsports.yahoo.com%2Farticles%2Fdan-piazza-title-defense-national-224856350.html&c=14632325196306918722&mkt=en-usThe 27th National Thoroughbred Racing Association (NTRA) National Horseplayers Championship (NHC) presented by Caesars Sportsbook, Horseshoe Las Vegas, and Racetrack Television Network (RTN) kicks off ...Mar 6, 2026 2:48 PM
i-am-neon.itch.io
Show HN: Voiced, image-based D&D inspired AI-native RPGhttps://i-am-neon.itch.io/infinitI'm a solo dev and I built a visual novel-style RPG where you type what you want to do and an AI game master responds in real time. Free alpha, plays in the browser.What makes it different from AI Dungeon: the AI doesn't just generate text. It emits structured commands that change the music, move NPCs between locations, give/remove items, swap character portraits based on emotional reactions, and trigger cutscenes. Cinematic stills are generated on the fly with Flux 2 Klein 4B, and characters are voiced in real time via Inworld. Separate AI agents maintain a quest journal and write save summaries. The result feels more like a tabletop RPG session than a chatbot conversation.The world is hand-crafted, not AI-generated. I wrote all the locations, characters, and lore by hand (Himalayan fantasy setting inspired by travel through Nepal and Bhutan). The AI's job is to run the game inside that authored world. Everyone explores the same world, every playthrough is different.Stack: Godot 4.5 cMar 6, 2026 2:43 PM
swipeforsex.com
Show HN: Swipeforsex.com – an experiment in honesty in dating appshttps://www.swipeforsex.com/I bought the domain swipeforsex.com mostly because I thought the name was funny.Instead of parking or selling it, I decided to build a small experiment around it: a dating site that removes the usual pretending and focuses on what many dating apps quietly optimize for anyway.The idea was simple: what happens if you remove the euphemisms and make the intent explicit?The project is intentionally minimal. I built it quickly just to test whether people would actually use something this direct or if the entire premise would fail.I'm mostly curious about two things:1. Whether the blunt positioning changes user behavior compared to typical dating apps. 2. Whether the domain itself drives curiosity traffic.Would love feedback from HN about the idea, UX, or whether this is a terrible concept.Happy to answer any technical or product questions.Mar 6, 2026 2:27 PM
github.com
Show HN: Own-term – run your developer portfolio in the terminal (npx own-term)https://github.com/Biki-dev/own-termHi HN! I'm Biki, the author.Own-term is a TypeScript framework that lets developers create interactive terminal portfolios, runnable by anyone with Node.js via `npx own-term`.The idea: instead of a portfolio website, your portfolio is a CLI experience — with commands like `about`, `projects`, `skills`, `contact`. You configure it with a TypeScript config file and can distribute it on npm.Technically it uses inquirer for the interactive prompts, chalk + gradient-string for terminal colors, and a simple plugin API that lets you register custom commands.The project is MIT licensed and actively looking for contributors — several good-first-issues are open covering test coverage, live theme switching, and tab completion.Happy to answer any questions about the architecture or design decisions.Mar 6, 2026 2:24 PM
github.com
Show HN: Moongate – Ultima Online server emulator in .NET 10 with Lua scriptinghttps://github.com/moongate-community/moongatev2I've been building a modern Ultima Online server emulator from scratch. It's not feature-complete (no combat, no skills yet), but the foundation is solid and I wanted to share it early.What it does today: - Full packet layer for the classic UO client (login, movement, items, mobiles) - Lua scripting for item behaviors (double-click a potion, open a door — all defined in Lua, no C# recompile) - Spatial world partitioned into sectors with delta sync (only sends packets for new sectors when crossing boundaries) - Snapshot-based persistence with MessagePack - Source generators for automatic DI wiring, packet handler registration, and Lua module exposure - NativeAOT support — the server compiles to a single native binary - Embedded HTTP admin API + React management UI - Auto-generated doors from map statics (same algorithm as ModernUO/RunUO)Tech stack: .NET 10, NativeAOT, NLua, MessagePack, DryIoc, KestrelWhat's missing: Combat, skills, weather integration, NPC AI. This is still early — theMar 6, 2026 2:22 PM
conutil.com
Show HN: I made a privacy-first browser-based image compressorhttps://www.conutil.com/Hey HN.I’m a 17-year-old developer, and I recently needed to compress a lot of images for a small website I was working on.Most tools I found required uploading files to their servers.That made me uncomfortable, especially since many of the images belonged to clients.I also ran into another issue: many tools were slow at compressing multiple images because of server uploads.So I decided to build something that works differently.I built ConUtil, an image toolkit that runs entirely in the browser.AhmedMar 6, 2026 2:10 PM

spacenews.com
Hyperscalers are coming to an orbit near you. Power will decide the winners.https://spacenews.com/hyperscalers-are-coming-to-an-orbit-near-you-power-will-decide-the-winners/Amid the explosive growth surrounding telecommunications megaconstellations, orbital data centers and next-generation payloads, the space ecosystem is entering a period of rapid and irreversible change. Announcements and filings for satellite constellations numbering in the tens of thousands, hundreds of thousands and now even 1 million-plus are becoming commonplace. The waves that even a fraction of […] The post Hyperscalers are coming to an orbit near you. Power will decide the winners. appeared first on SpaceNews.Mar 6, 2026 2:00 PM
news.ycombinator.com
Ask HN: Do You Have a Homelab?https://news.ycombinator.com/item?id=47274800I'm Ricard Bejarano, and together with O'Reilly, I'm writing The Homelab Handbook, the definitive guide to homelabbing and self-hosting.To inspire readers, we want the last chapter to be a series of real world homelab examples, to show there's not one prescription for what a homelab is—a homelab is what you make it. As such, we're looking for homelabbers that would like to have their homelabs featured in the book. We're looking for variety, in hardware, software, and scale—from a single Raspberry Pi, to full-height racks in the basement—so don't be shy to share yours.If you'd like to submit yours to be featured in the book, please complete the following form:https://docs.google.com/forms/d/e/1FAIpQLScLc16kBFDY3liuEz_a40CF1sYz7yeqPmy1CKVhufHNSzjhIA/viewformForm submission deadline is Apr 12th, 2026. We will reach out to all submitters shortly after that with a response on whether your homelab was selected or not. If we accept yours, you will be asked to share more details so I can writeMar 6, 2026 1:43 PM