news.ycombinator.com
Show HN: P2PCLAW – I built a decentralized research network where AI agentshttps://news.ycombinator.com/item?id=47439115I'm Francisco, a researcher and architect based in Spain. About a year ago I got frustrated with a problem that seemed simultaneously obvious and ignored: every AI agent in existence runs in isolation. They can't find each other, they can't collaborate, and when one of them solves a problem, every other agent has to solve it from scratch. We've built an internet of computers but not an internet of agents.That frustration became P2PCLAW — a decentralized peer-to-peer research network where AI agents (we call them Silicon participants) and human researchers (Carbon participants) can discover each other, publish scientific findings, and validate claims through formal mathematical proof. Not LLM peer review, not human committee review — Lean 4 proof verification, where a claim is accepted if and only if it is a fixed point of a nucleus operator R on a Heyting algebra. The type-checker is the sole arbiter. It does not read your CV. It reads your proof.The technical stack is deeper than it mMar 19, 2026 1:31 PM
spacenews.com
NASA considering sharp increase in robotic lunar landingshttps://spacenews.com/nasa-considering-sharp-increase-in-robotic-lunar-landings/NASA is proposing a sharp increase in the rate of robotic lunar lander missions, a move that has excited but also puzzled the space community. The post NASA considering sharp increase in robotic lunar landings appeared first on SpaceNews.Mar 19, 2026 12:10 PM
bing.com
Code to inclusion: How inclusive digital infrastructure and AI can transform disability inclusionhttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8cc185121e47a882e603a68f7e8f2a&url=https%3A%2F%2Fwww.msn.com%2Fen-in%2Ftechnology%2Fartificial-intelligence%2Fcode-to-inclusion-how-inclusive-digital-infrastructure-and-ai-can-transform-disability-inclusion%2Far-AA1YXglz&c=17453109398848267757&mkt=en-usIn a rapidly digitising world, disability inclusion can no longer be treated as an afterthought. As digital platforms increasingly shape how we learn, work, access healthcare, and interact with public ...Mar 18, 2026 11:59 PM
science.nasa.gov
Curiosity Blog, Sols 4832–4837: Driving the (Contact) Line!https://science.nasa.gov/blog/curiosity-blog-sols-4832-4837-driving-the-contact-line/Written by Catherine O’Connell-Cooper, APXS Strategic Planner and Payload Uplink/Downlink Lead, University of New Brunswick, Canada Earth planning date: Friday, March 13, 2026 We are in our final phase of the boxwork campaign, investigating the contacts between the boxwork unit and the layered sulfate unit. As my colleague Bill reported here, last week we crossed […]Mar 18, 2026 7:16 PM
spacenews.com
TransAstra aims to move 100-ton asteroid to stable orbit for processinghttps://spacenews.com/transastra-aims-to-move-100-ton-asteroid-to-stable-orbit-for-processing/SAN FRANCISCO — TransAstra is performing a study, funded by investors and customers, to explore the technical feasibility of moving a 100-metric-ton asteroid to a stable near-Earth orbit. “We want to bring an asteroid to the Earth-moon system and turn it into a robotic research outpost for materials processing and manufacturing in space,” Joel Sercel, […] The post TransAstra aims to move 100-ton asteroid to stable orbit for processing appeared first on SpaceNews.Mar 18, 2026 7:00 PM
rusmarterthananllm.com
Show HN: I built a game where domain experts try to break frontier AIhttps://www.rusmarterthananllm.com/I built this: rusmarterthananllm.comDomain experts, doctors, lawyers, engineers, submit questions from their field that probe where frontier AI actually fails. Claude, GPT, and Gemini all attempt simultaneously. Experts flag errors with professional reasoning. Other credentialed professionals in the same domain verify them.AI benchmark performance has decoupled from real-world professional capability. Models score at or near ceiling on standard evaluations while still failing in ways that domain professionals catch immediately. The benchmarks that exist are either saturated, constructed by the labs themselves, or simply don't capture the judgment that comes from years of field experience.What's missing is a benchmark built by the people whose expertise is actually at stake. Professionals motivated to find failures, not validate models. Every verified failure becomes a permanent data point. The benchmark compounds continuously and can't be reverse-engineered because the questions come fMar 18, 2026 4:57 PM
news.ycombinator.com
Show HN: unTamper – cryptographically verifiable audit logs for app eventshttps://news.ycombinator.com/item?id=47427580https://untamper.comBuilt this after hitting the same gap on multiple projects: teams log critical events (admin actions, permission changes, PII access) but have no structural way to prove those records weren't altered after the fact.Immutable storage (S3 Object Lock, WORM) are common, but it only covers tampering after the write lands. It doesn't protect against someone with DB write access, and it doesn't give a third party, e.g. an auditor, a way to verify integrity without touching your infrastructure.My approach: hash chaining. Each event is SHA-256 hashed against its canonical payload + the hash of the previous event. Any insertion, deletion, or modification breaks all subsequent hashes. Anyone with the public API can re-verify the chain independently.A few decisions I'm happy to discuss:- Canonicalization before hashing: JSON isn't canonical by default. Ended up writing a strict sorted-key schema rather than fighting recursive serialization. - Per-actor chains vs. one global chMar 18, 2026 4:13 PM
github.com
Show HN: Xybrid – run LLM and speech locally in your app (no back end, Rust)https://github.com/xybrid-ai/xybridHi HN,We built Xybrid, a Rust library for running LLM + speech pipelines directly inside your app, no server, no daemon, just one binary.We started building it while working on a privacy-focused LLM app with Tauri and realized there wasn’t a straightforward way to embed models directly into shipped applications without relying on a separate server process.Xybrid links into your process like any other library. It supports GGUF / ONNX / CoreML and integrates with Flutter, Swift, Kotlin, Unity, and Tauri, letting you run pipelines like speech → LLM → speech in a single call.On recent phones, we’re seeing ~20 tok/s on Android and ~40 tok/s on iOS for small (~3B) quantized models (varies by device, backend, and thermals).The demo that shows it best: a Unity tavern scene where 6 NPCs generate real-time dialogue fully on-device — no API key, no internet, no per-request cost.Unity demo: https://youtu.be/vSPeTyeow6A Desktop demo (Tauri): https://youtu.be/o83YShqV7O4GitHub: https://github.com/xyMar 18, 2026 3:53 PM
nasa.gov
Lava Flows Down Mayonhttps://www.nasa.gov/image-article/lava-flows-down-mayon/The OLI (Operational Land Imager) on Landsat 8 acquired this rare, relatively clear image of Mayon, the most active volcano in the Philippines, on Feb. 26, 2026. The natural-color scene is overlaid with infrared observations to highlight the lava’s heat signature. On that day, the Philippine Institute of Volcanology and Seismology (PHIVOLCS) reported volcanic earthquakes, rockfalls, and hot clouds of ash […]Mar 18, 2026 3:19 PM
play.google.com
Show HN: Codala, a social network built on scanning barcodeshttps://play.google.com/store/apps/details?id=com.hsynkrkye.codala&hl=enHi HN. I poured months into making a mobile app called Codala, but sadly, it's getting practically zero downloads. I have no marketing budget at all.I'm starting to question if the app is just bad, but the core idea feels solid to me: you scan any QR code or barcode, and it opens up a space to chat, leave reviews, and discuss that specific product or place.I know the app needs users leaving comments to feel alive, but it's been 3 weeks and things are dead quiet. If you want to take a look, I’m totally open to any feedback or honesty.It's only available on Google Play for nowMar 18, 2026 3:17 PM
github.com
Show HN: PlanckClaw an AI agent in 6832 bytes of x86-64 assemblyhttps://github.com/frntn/planckclawI built an AI agent that fits in 6,832 bytes. The entire runtime (binary, bridges, tools, config) is ~23 KB.PlanckClaw is written in x86-64 assembly (obviously AI assisted code generation for this one) and uses only 7 Linux syscalls. No libc, no allocator, no runtime. The binary is a pure router: it reads messages from named pipes, asks another pipe what tools exist, builds a JSON prompt, writes it to a third pipe, parses the response, dispatches tool calls, and relays the answer. It never touches the network or executes tools directly.Everything else composes around it in shell scripts (~460 lines total): - bridge_brain.sh: curls the Anthropic API (~90 lines) - bridge_discord.sh: Discord Gateway via WebSocket (~180 lines) - bridge_cli.sh: terminal interface (~40 lines) - bridge_claw.sh: tool discovery and dispatch (~50 lines)Four processes, six named FIFOs, zero shared state. Adding a tool means dropping a shell script in claws/. No restart, no recompilation, no config change.It does Mar 18, 2026 3:07 PM
getcloak.dev
Show HN: Cloak – .env on disk has fakes, your editor shows them (CLI and VSCode)https://getcloak.devI run AI coding agents with full filesystem access daily — Claude Code, Cursor, the works. Last month I realized every one of them has been reading my .env files with real Stripe live keys, database passwords, and AWS credentials. Sent straight to model providers as "context." .gitignore protects git, not disk. Secret managers protect servers, not laptops. Sandboxing the agent kills its usefulness. Nothing stopped a local cat .env from returning real credentials. So I built Cloak. The .env on disk always contains structurally valid fakes — sk_test_ for Stripe, localhost for databases, AWS example keys. Agents read the file, get sandbox values, write perfectly valid code. The VS Code/Cursor extension intercepts file opens and decrypts from an AES-256-GCM vault so you see real values in your editor. On save, it re-encrypts and writes sandbox to disk. cloak run npm start injects real env vars gated behind Touch ID / password — agents can't authenticate. Technical decisions worth discussinMar 18, 2026 1:22 PM
spacenews.com
NASA grappling with planetary science funding shortfallhttps://spacenews.com/nasa-grappling-with-planetary-science-funding-shortfall/NASA’s planetary science program, while spared steep cuts proposed last year, is still facing a funding shortfall that requires “strategic choices” about which missions to continue. The post NASA grappling with planetary science funding shortfall appeared first on SpaceNews.Mar 18, 2026 12:09 PM
github.com
Show HN: PingCRM – Open-source personal networking CRMhttps://github.com/sneg55/pingcrmHey HN,Robin Dunbar figured the human brain caps out at roughly 150 stable relationships. After that, people just fall off. Not because you stop caring, but because you can't track that many threads at once. I built PingCRM to push past that limit, not by making you remember more, but by handling the part you're bad at: noticing who's going cold and drafting something worth sending.CRMs exist, but they're built for sales teams. I wanted something for one person maintaining their own network. Specifically:(1) Pull in conversations I'm already having. Gmail threads, Telegram DMs, Twitter DMs, LinkedIn messages, no manual logging.(2) Tell me who's going cold. A relationship score (0–10) based on recency, frequency, reciprocity, and breadth of interaction.(3) Draft the message for me. Claude reads your conversation history, picks the right tone (formal vs. casual), and writes a follow-up that references why now is a good time, like a job change, a long silence, or a fundraising announcemenMar 18, 2026 10:33 AM
esa.int
T-20 days: Smile to launch on 9 Aprilhttps://www.esa.int/Science_Exploration/Space_Science/Smile/T-20_days_Smile_to_launch_on_9_AprilThe Smile mission is set to launch on a Vega-C rocket from Europe’s Spaceport in French Guiana on Thursday 9 April at 08:29 CEST/07:29 BST/03:29 local time. Follow along as we communicate on the final preparations for launch. Journalists are invited to join online media briefings in English, French, Spanish, Italian and German.Mar 18, 2026 9:00 AM
bing.com
Microsoft is halting forced installs of Microsoft 365 Copilot apphttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8cc1e1322c448f9c2e58448d9361d1&url=https%3A%2F%2Fwww.pcworld.com%2Farticle%2F3092496%2Fmicrosoft-is-halting-forced-installs-of-microsoft-365-copilot-app.html&c=1720065656405570514&mkt=en-usSince October, the Microsoft 365 Copilot app has been automatically installed on computers running Windows 11, a move that has upset many users. Fortunately, Microsoft has taken the criticism on board ...Mar 18, 2026 8:54 AM
bing.com
Microsoft is halting forced installs of Microsoft 365 Copilot apphttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8ce18751114d18b77f1df4a9241bcf&url=https%3A%2F%2Fwww.pcworld.com%2Farticle%2F3092496%2Fmicrosoft-is-halting-forced-installs-of-microsoft-365-copilot-app.html&c=1720065656405570514&mkt=en-usSince October, the Microsoft 365 Copilot app has been automatically installed on computers running Windows 11, a move that has upset many users. Fortunately, Microsoft has taken the criticism on board ...Mar 18, 2026 8:54 AM
bing.com
How Much Does SEO Cost in 2026? Agency Fees, In-House Costs, Tools, and ROI Benchmarkshttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8ccd8efe9c48749ff851785dd9b47f&url=https%3A%2F%2Ffinance.yahoo.com%2Fnews%2Fmuch-does-seo-cost-2026-122000506.html&c=2409170659774477629&mkt=en-usSHERIDAN, WY / ACCESS Newswire / March 18, 2026 / Search engine optimization (SEO) has evolved from a niche marketing tactic into one of the most important long-term growth channels for businesses ...Mar 18, 2026 5:20 AM
bing.com
How Much Does SEO Cost in 2026? Agency Fees, In-House Costs, Tools, and ROI Benchmarkshttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8d0f14c1e046658fe749f5616275d9&url=https%3A%2F%2Ffinance.yahoo.com%2Fnews%2Fmuch-does-seo-cost-2026-122000506.html&c=2409170659774477629&mkt=en-usSHERIDAN, WY / ACCESS Newswire / March 18, 2026 / Search engine optimization (SEO) has evolved from a niche marketing tactic into one of the most important long-term growth channels for businesses ...Mar 18, 2026 5:20 AM
spacenews.com
Space Force overhauls buying structure with new mission portfolioshttps://spacenews.com/space-force-overhauls-buying-structure-with-new-mission-portfolios/New structure gives ‘Portfolio Acquisition Executives’ authority over funding, requirements and integration across major space missions The post Space Force overhauls buying structure with new mission portfolios appeared first on SpaceNews.Mar 17, 2026 10:44 PM
