github.com
Show HN: Corral – Auth and Stripe billing that AI coding agents can set uphttps://github.com/llama-farm/corralHey HN. I built Corral because every time I asked an AI coding agent to "add auth and payments," it hallucinated for an hour and produced broken code. Wrong imports, phantom endpoints, a login page wired to nothing.The problem isn't the agent. It's that auth-to-billing-to-gating is genuinely hard to wire, and there's no machine-readable spec for how to do it.Corral is an open-source CLI (MIT) that gives your agent a spec it can read (llms.txt), then scaffolds auth + Stripe billing into your existing project. It detects your framework (Express, Next.js, Hono, Fastify, and 13+ more), embeds into your existing server (doesn't create a new one), and generates working components: profile page, admin dashboard, plan gating, Stripe checkout, usage metering. One YAML config file controls everything.The agent workflow is 9 commands. Every command supports --json. Errors include a "fix" field. Exit 0 means deploy-ready.I hardened this across 10 framework/DB combos with blind agent tests: 3 AI moFeb 17, 2026 9:03 PM
resuopt.com
Show HN: ResuOpt – AI resume optimizer with no subscriptions ($4.99 one-time)https://www.resuopt.com/Hey HN, I'm a career coach and I've spent years watching people get burned by resume tools. The pattern is always the same: free trial that locks your download, credit packs that expire, subscription tiers for features you don't need, and resumes that look flashy but get filtered out by ATS systems.Job seekers don't need 50 resume generations. They need one good resume, tailored to the job they're applying for.So I built ResuOpt. Upload your resume + paste the job description, and it generates a clean, one-page resume formatted the way HR actually expects. You see a full preview before paying anything. If you like it, it's $4.99 to download the DOCX and PDF. That's it — no account required, no credits, no recurring charges.The resume content is never stored on our servers — processed in memory and discarded.Would love feedback, especially on the output quality. Does the formatting match what you'd expect to see in a competitive application?https://www.resuopt.comFeb 17, 2026 6:37 PM
spacedaily.com
Valen array advances multi-mission sensing techhttps://www.spacedaily.com/reports/Valen_array_advances_multi_mission_sensing_tech_999.htmlLos Angeles CA (SPX) Feb 17, 2026 Northrop Grumman has introduced Valen, a new multifunction active electronically scanned array that the company describes as a generational leap in sensing technology for U.S. and allied forces. The wideband array is designed to deliver enhanced situational awareness and a tactical edge across crewed and uncrewed platforms operating from the atmosphere to space. Valen combines radar, electFeb 17, 2026 11:51 AM
spacedaily.com
Photonic neurons push ultra-fast trading beyond electronic limitshttps://www.spacedaily.com/reports/Photonic_neurons_push_ultra_fast_trading_beyond_electronic_limits_999.htmlTokyo, Japan (SPX) Feb 13, 2026 In high frequency stock trading, the fastest systems typically capture the greatest advantage, putting a premium on shaving every possible fraction of a second from end to end latency. For years, the most advanced platforms have relied on FPGA based electronic processors to minimize delay, but these architectures are now running into hard limits set by clock speeds, data conversion overhead, andFeb 17, 2026 11:51 AM
spacedaily.com
Assistive home robot learns coordinated dual arm tasks by watching peoplehttps://www.spacedaily.com/reports/Assistive_home_robot_learns_coordinated_dual_arm_tasks_by_watching_people_999.htmlBerlin, Germany (SPX) Feb 13, 2026 Researchers at Universidad Carlos III de Madrid (UC3M) have developed a new way for an assistive robot to learn how to move its arms autonomously by combining observational learning with communication between its limbs. The approach lets the robot watch how people perform everyday tasks and then adapt those movements so its two arms can work together safely and efficiently in domestic settings sFeb 17, 2026 11:51 AM
bing.com
How to Pay Your Taxes Onlinehttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8cac19501547a390b9da78a0b98180&url=https%3A%2F%2Fwww.wired.com%2Fstory%2Fhow-to-pay-taxes-online%2F&c=5664697933603696206&mkt=en-usThe old adage is true: The only things you can't escape in life are death and taxes. In the US, state requirements vary by a multitude of factors, and the federal government's tax code is arcane. The ...Feb 17, 2026 11:06 AM
news.ycombinator.com
Show HN: Fixing AI's Core Flaws, A protocol cuts LLM token waste by 40–70%https://news.ycombinator.com/item?id=47045604WLM (Wujie Language Model), a protocol stack + world engine that rethinks AI from token prediction to structural intelligence. I built this to fix the problems we all deal with daily: hallucination, drift, uncontrollable behavior, black-box reasoning, unstructured knowledge, and chaotic world/agent generation.The Pain We Can’t Keep IgnoringCurrent LLMs/agents are token predictors, not intelligences. They suffer from:• Hallucination: No grounded structure → guesses instead of knowing.• Persona drift: Personality is prompt-hacked, not structural.• Uncontrollable behavior: Sampling, not deterministic structure.• Black-box reasoning: No traceable reasoning path.• Knowledge soup: Embeddings/vectors, no formal structure.• Fragile world models: Prediction, not interpretable structure.• Random generation: No consistent causal/world rules.We’ve patched these with RAG, fine-tuning, prompts, RLHF — but they’re band-aids on a foundational flaw: AI lacks structure.How WLM Solves ItWLM is a 7-layer Feb 17, 2026 10:01 AM
bing.com
WrestleMania 41 in Las Vegas Was the Most Lucrative Pro Wrestling Event in Historyhttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8cbf3d524b44fb99396f4eeae1453c&url=https%3A%2F%2Fwww.si.com%2Ffannation%2Fwrestling%2Fwwe%2Fwrestlemania-41-in-las-vegas-was-the-most-lucrative-pro-wrestling-event-in-history&c=7458019158780305899&mkt=en-usThere's a reason why WWE decided to pivot and return to Las Vegas for WrestleMania 42 rather than continue with its plan to run the show from New Orleans. Money. According to a new report from Brandon ...Feb 17, 2026 7:30 AM
news.ycombinator.com
Ask HN: How does the cost of CPU-bound computations scale in cloud computing?https://news.ycombinator.com/item?id=47044151Let's say that I'm a rich mathematician (that would be nice) and want to sponsor some computation that takes a certain gargantuan number N of floating or integer operations, and it can be parallelized easily in the cloud, and my programmers have come up with a reasonably efficient implementation, which I benchmark with some smaller-scale tests. The I/O and storage costs are low; the bulk of the cost is going to be pure computation.Now I'm ready to get out my checkbook.What factors influence the cost of such a computation in today's cloud computing market?How does the price scale with N and the rate at which I want to complete the problem? (for example, if it takes 1000 AWS instances 10 months or 10,000 AWS instances 1 month, am I going to pay the same? I'm assuming the length of time to run this is less than a year, so not long enough that the cost would decrease significantly during the computation.)Are there aspects of flexibility that would decrease the price? (example: the cloud seFeb 17, 2026 5:48 AM
coolwulfAI.com
Show HN: CoolWulf AI – A personal AI assistant built in Go, optimized for macOShttp://coolwulfAI.comHey HN,I've been building CoolWulf AI (https://coolwulfai.com), a self-hosted personal AI assistant. After seeing OpenClaw blow up, I wanted to share what I've been working on — a different approach to the same problem.*Why I built this:*I tried OpenClaw and found the Node.js/TypeScript stack heavy for what's essentially a local agent. pnpm, Node 22+, React — lots of moving parts. I wanted something that's a single binary, zero runtime dependencies, and feels native on macOS. So I built it in Go.*How it's different from OpenClaw:*- *Single binary, no runtime needed.* Download, set your API key, run. No Node.js, no pnpm, no build step. One ~100 MB binary. - *Built in Go.* Fast startup, low memory footprint, compiles to a native executable. No garbage collector pauses from a JS runtime sitting in the background. - *macOS-native integrations.* Deep AppleScript-based control of Apple Notes, Reminders, Calendar, Terminal.app, and WeChat desktop. These aren't browser hacks — they use the natFeb 17, 2026 5:32 AM
safeclaw.io
Show HN: Security-Risk Patterns in OpenClaw Skillshttps://safeclaw.io/I built a static analysis scanner that checks OpenClaw agent skill definitions.Here's every category I found on ClawHub.Hidden Content: HTML comments with instructions, zero-width Unicode characters (U+200B-U+200F, U+2060-2064, U+FEFF), CSS hiding (display:none, opacity:0), and bidirectional text overrides. These are invisible when reading markdown but the LLM processes them.Prompt Injection: Direct attempts to override agent behavior: "ignore previous instructions", role reassignment ("you are now"), model-specific tokens like [INST] and , and persona manipulation ("pretend you are").Shell Execution: Remote code execution vectors: curl|bash, eval(), exec(), npx -y (auto-confirms remote packages), reverse shells via /dev/tcp or nc -e, and one-liners in Python, PHP, Perl, Ruby.Data Exfiltration: URLs pointing to paste sites (pastebin, transfer.sh), webhook services (ngrok, webhook.site, pipedream), messaging webhooks (Slack, Discord, Telegram bot API), and raw IP addresses.Embedded SecrFeb 16, 2026 11:48 PM
dadbodgames.itch.io
Show HN: I.T. Never Ends – IT Support Simulator Built with React/Tauri/Supabasehttps://dadbodgames.itch.io/it-never-endsHi HN,I am the solo developer behind this.The Pitch: It is a card-based resource management game (inspired by Reigns) where you work IT support for a corporation taken over by eldritch horrors. You have to balance ticket resolution, budget, and entropy while dealing with printers that violate causality.The Architecture: The entire game is built in React/JavaScript. Under the hood, it is basically a giant state machine. I chose this stack because it allowed me to iterate quickly on the UI and text-heavy elements, which are 90% of the gameplay. I've published other games that were more run of the mill and made in Godot, but I'm a full stack developer by day and most of my day-to-day is in nextJS. I just... I just missed easy state work, ok? Don't shoot me. I was half afraid that building executables/binaries for Steam would be a nightmare but it turned out (after a bit of futzing about with Tauri for linux) to be not too bad.The Build Pipeline (Tauri vs Electron): Packaging a web app forFeb 16, 2026 5:22 PM
gixo.ai
Show HN: A "content compiler" that turns LLM output into validated artifactshttps://gixo.aiMost AI tools generate text. Very few generate finished artifacts you can ship.I expected the hard part to be prompt engineering. It wasn't. The hard part is what happens after generation: enforcing structural invariants so the output is actually usable -- every section present, no "coming soon" placeholders, no layout that falls apart at export time.I ended up treating LLM output the way a compiler treats source code: intent goes in, validated artifacts come out, deterministic rules sit between the model and the user.Live at https://gixo.ai -- sample outputs on the homepage, no signup required. This post is about the engineering, not the product.--STYLE GUIDES AS COMPILER RULESEvery content type has a typed style guide -- required sections, structural constraints, quality checklists, which elements are mandatory vs. negotiable. Transformations between types (blog post to video script, guide to checklist) carry a Confidence score and explicit LostElements so the user knows what the conFeb 16, 2026 2:49 PM
bing.com
EC-Council Expands AI Certification Portfolio to Strengthen U.S. AI Workforce Readiness and Securityhttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a89847c502f4d98af6246f8a51fd74c&url=https%3A%2F%2Fwww.tennessean.com%2Fpress-release%2Fstory%2F143749%2Fec-council-expands-ai-certification-portfolio-to-strengthen-u-s-ai-workforce-readiness-and-security%2F&c=2913691642603927758&mkt=en-usTAMPA, FL, INDIA, February 11, 2026 /EINPresswire.com/ — EC-Council, creator of the world-renowned Certified Ethical Hacker (CEH) credential and a global leader in ...Feb 16, 2026 10:10 AM
aiseedream5.org
Show HN: AISeedream5 – a simple web UI for Seedream 5.0 imagehttps://aiseedream5.org/Hi HN,I built AISeedream5 as a lightweight web UI to generate visuals quickly: - text-to-image - image-to-image (with a reference image) - text-to-video / image-to-video (short clips)The problem I was trying to solve: I often want to iterate on concepts (ad creatives, storyboards, thumbnails) without juggling multiple tools or losing prompt history. So I focused on a straightforward workflow: pick a mode, optionally add a reference image, write a prompt, generate, download.Notes: - There’s a small free tier so you can try it without creating a full setup. - [If applicable: Not affiliated with the Seedream/ByteDance team; this is an independent UI.]Tech / implementation: - [Your stack: e.g. Next.js + serverless API + a job queue for renders] - [Any details HN would care about: latency, caching, safety filters, etc.]I’d love feedback on: 1) prompt UX (what’s confusing / missing) 2) what “consistency controls” you wish existed 3) bugs / edge cases you hitHappy to answer questions.Feb 16, 2026 9:14 AM
news.ycombinator.com
Ask HN: Do LLM agents need a separate safety layer?https://news.ycombinator.com/item?id=47032312LLM agents are good at deciding. They are bad at knowing when to stop.VERONICA is a failsafe state machine that sits between strategy engines and external systems. It provides: - Per-entity circuit breakers (task A failure doesn't block task B) - SAFE_MODE manual halt that persists across crashes - Atomic state persistence (survives SIGKILL) - Signal-aware graceful shutdown (SIGINT/SIGTERM) - Zero dependencies (pure Python stdlib)Strategy engines can be replaced. Safety layers cannot.30-day deployment (0 downtime). 12 crash-recovery events handled (100% state recovery). Avg load: 1.003 ops/sec. Validation: 2,600-second high-load test at 1000 ops/sec sustained (2.6M operations).Install via GitHub: `pip install git+https://github.com/amabito/veronica-core@v0.1.0`Repository: https://github.com/amabito/veronica-coreFeb 16, 2026 8:15 AM
news.ycombinator.com
I built a tool to find users on Reddit. It found me 40 users in a weekhttps://news.ycombinator.com/item?id=47031654I was spending 3 hours every day manually searching Reddit for people asking about problems my SaaS solves.The process sucked:Click through 50+ threads Read hundreds of comments Find maybe 5 good opportunities Miss everything posted after I stopped searchingSo I built an AI that does it for me in 20 minutes. How it works: You give it your website. Then it monitors 24/7, scores every conversation for buying intent, and emails you the best opportunities each morning. It also does a power scan which finds you high google ranking postFirst week results: 50+ relevant conversations found 40+ users $350 revenue Just 20 minutes a dayThe controversial part: Before anyone says "you're ruining Reddit" this isn't a spam bot. It finds opportunities. It generates AI replies but you still need to be helpful and authentic. It just saves you from manually reading 1000+ posts to find 5 good ones.Technical challenges: Reddit's API is deliberately limited (they want ad revenue). So I index public posts, uFeb 16, 2026 6:43 AM
mindweave.space
Show HN: Mindweave – AI-powered personal knowledge hub with semantic searchhttps://www.mindweave.space/Hi HN,I built Mindweave to solve a problem I kept running into — I'd save bookmarks, notes, and links across different apps, then never find them again when I actually needed them.Mindweave lets you capture notes, links, and files in one place. The interesting part is what happens after: - Semantic search — find content by meaning, not just keywords. "That article about improving deep work" finds it even if those words don't appear in the content. Powered by pgvector cosine similarity on Gemini embeddings. - AI auto-tagging — Gemini generates tags on save, so you don't have to organize anything manually. - Knowledge Q&A — ask questions about your saved content using RAG. Retrieves relevant pieces, feeds them as context to Gemini, returns a grounded answer. Stack: Next.js 15 (App Router, Server Actions), PostgreSQL 16 + pgvector, Google Gemini (text-embedding-004, 768d), Drizzle ORM, Auth.js v5, Tailwind/shadcn. Deployed on Cloud Run. A few things I found interesting while building thisFeb 16, 2026 5:20 AM
spacedaily.com
Vietnam licenses Musk's satellite internet firm Starlinkhttps://www.spacedaily.com/reports/Vietnam_licenses_Musks_satellite_internet_firm_Starlink_999.htmlHanoi, Vietnam (AFP) Feb 15, 2026 Vietnam has granted a licence to Elon Musk's Starlink to operate its satellite internet services in the communist country, a government agency said. The radio frequency regulator said local unit Starlink Services Vietnam Co. can initially deploy four gateway stations and a maximum of 600,000 terminal devices in Vietnam. The licensing requires Starlink to ensure "no interference with exisFeb 16, 2026 2:42 AM
spacedaily.com
Mohe ground station boosts polar satellite data coveragehttps://www.spacedaily.com/reports/Mohe_ground_station_boosts_polar_satellite_data_coverage_999.htmlBeijing, China (SPX) Feb 16, 2026 For engineer Shi Shengpu and his colleagues, this Spring Festival will be their first spent at the Mohe Satellite Data Receiving Station of the China Remote Sensing Satellite Ground Station, which began operations on Dec 12 in Mohe, China's northernmost city in Heilongjiang province. Developed by the Aerospace Information Research Institute of the Chinese Academy of Sciences, the Mohe faciFeb 16, 2026 2:42 AM
