news.ycombinator.com
Show HN: Voicetest – open-source test harness for voice AI agentshttps://news.ycombinator.com/item?id=47048811We've been building voice agents across Retell, VAPI, LiveKit, and Bland, and the testing story is... rough. Every platform has its own config format, there's no shared way to define what "correct" looks like, and most teams end up doing manual QA by literally calling their agent and listening. So we built voicetest.voicetest is an open source (Apache 2.0) test harness that works across voice AI platforms. You import your agent graph from any supported platform (or define one from scratch), write test scenarios with expected behaviors, and voicetest simulates conversations and evaluates them with LLM judges that score each turn 0.0-1.0 with written reasoning. It also ships global compliance evaluators for things like HIPAA, PCI-DSS, and brand voice consistency. The core abstraction is an AgentGraph IR that normalizes across platform formats, so you can convert between Retell, VAPI, LiveKit, and Bland configs and test them all the same way.Quick start:``` uv tool install voicetest voiceFeb 17, 2026 3:49 PM
github.com
Show HN: AudioNimbus – Safe Rust Wrapper for Steam Audiohttps://github.com/MaxenceMaire/audionimbus/releases/tag/0.12.0I just released a new version of AudioNimbus. AudioNimbus is a safe Rust wrapper around Steam Audio, Valve's library for physics-based spatial audio (occlusion, reflections, reverb, HRTF) in games.v0.12.0 comes with support for the latest Steam Audio version, major safety improvements, improved API ergonomics and documentation.The library also includes an 'auto-install' feature that automatically downloads and links Steam Audio for you, removing the usual hassle of linking a C library.Link to the project: https://github.com/MaxenceMaire/audionimbusIt is dual-licensed under MIT/Apache-2.0.Release notes: https://github.com/MaxenceMaire/audionimbus/releases/tag/0.1...Feb 17, 2026 2:52 PM
apps.apple.com
Show HN: Quick Issues: A Fast Mobile Issue Capture for GitHub, GitLab, and Giteahttps://apps.apple.com/us/app/quick-issues-fast-git-capture/id6758988655I got frustrated with how GitHub and GitLab's mobile apps handle issue creation. You need to be online, the flow is slow, and if you're on a self-hosted instance you're out of luck entirely.So I built Quick Issues. It's a lightweight Swift app with an offline buffer -- you capture the issue, it syncs when you have connectivity. Supports GitHub, GitLab, and Gitea/Forgejo (including self-hosted and local with PAT).A few technical notes for anyone interested: This was my first time using GRDB with SQLite instead of default Swift data structures, and the performance difference was significant. Setting up proper OAuth2 flows and GitHub/GitLab apps was more of an adventure than expected, but it's solid now.Free for a single account/instance, paid tier if you juggle multiple providers.My background is in GTD and data analytics, not traditional software engineering, so I'm genuinely curious: how does issue capture fit into your development workflow? Do you batch-create issues, capture them theFeb 17, 2026 2:50 PM
europeanspaceflight.com
Earth Observation Data Provider SatVu Closes £30 Million Funding Roundhttps://europeanspaceflight.com/earth-observation-data-provider-satvu-closes-30-million-funding-round/London-headquartered thermal infrared Earth observation data provider SatVu announced that it has closed a £30 million (approximately €34.3 million) funding round as it prepares to deploy a pair of satellites into orbit. The company launched its first satellite, HotSat-1, aboard a SpaceX Falcon 9 in June 2023. The satellite provided high-resolution thermal infrared imagery with […] The post Earth Observation Data Provider SatVu Closes £30 Million Funding Round appeared first on European Spaceflight.Feb 17, 2026 2:24 PM
seedance3ai.net
Show HN: Seedance3AI – a web app for text-to-video, image-to-videohttps://seedance3ai.net/Hi HN,I built Seedance3AI because I wanted a single, simple place to iterate on short visual ideas without hopping between tools.It currently supports: - text → video (short clips) - image → video (animate a reference image) - AI image generation (useful for thumbnails/storyboards)The workflow is intentionally minimal: pick a mode, set a couple of creative controls, generate, then download.Implementation notes: - [stack: e.g. Next.js + serverless API + a job queue + object storage] - [anything you learned the hard way: rate limits, retries, prompt history, etc.]What I’d love feedback on: 1) Which controls matter most (duration/aspect/seed/style, etc.) 2) Prompt UX: what’s confusing or missing 3) Any failure cases you hit (timeouts, weird outputs, etc.)[Optional: This is an independent project and not an official ByteDance/Seedance release.]Thanks!Feb 17, 2026 12:55 PM
bing.com
New Marriott Homes & Villas promotions: Save 10% on select rentals through April 5http://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8d9057a992480989bcc6dc7b533a3b&url=https%3A%2F%2Fwww.msn.com%2Fen-us%2Fmoney%2Fnews%2Fnew-marriott-homes-villas-promotions-up-to-10k-points-per-stay-25k-points-for-new-customers%2Far-AA1QeFIS&c=484492723698369850&mkt=en-usHomes & Villas by Marriott Bonvoy is the hotel giant’s answer to Airbnb, letting travelers book entire apartments, condos, homes, villas, and more. In short, they’re ideal for anyone traveling with a ...Feb 17, 2026 12:41 PM
bing.com
Hospital Staff Organized 'Beautiful' Wedding for Terminally Ill Bride Before She Died from Stage-4 Cancerhttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8d3a2b52a040358bb1a96a978e85d8&url=https%3A%2F%2Fwww.yahoo.com%2Flifestyle%2Farticles%2Fhospital-staff-organized-beautiful-wedding-201359791.html&c=10229408380979938841&mkt=en-usA terminally ill patient tied the knot shortly before her death, thanks to caregivers at a Colorado hospital. After marrying her husband Damon at a courthouse, Muranda "Mandy" Lambert had a formal ...Feb 17, 2026 12:13 PM
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
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
news.ycombinator.com
Ask HN: How's Business These Days for Upwork Freelancers?https://news.ycombinator.com/item?id=47045186I wonder if or how AI-assistance might be used more these days. By either the freelancers in delivering their services, or by buyers who might be figuring they don't need to hire a freelancer for whatever reason these days.Feb 17, 2026 8:54 AM
news.google.com
Best in Grass Lands in New York, Inviting 2,600 Consumers to Judge the State’s Top Cannabishttps://news.google.com/rss/articles/CBMihwFBVV95cUxNcEtFeVdCZkJwcnY1dXZKQWZYU3l6a0F5Uk9QYUVCVlVYZTlYMnBKajZ3am1Db181anU0VFpyWG5ROF92OEwtNUltbmRJaW5Kd0VMYXpIZXNubTZyZlUtaDlfdWwxeGJ2QXdQU3hSaXhTMm9XdUVRZl9aRDlBaEZBdmNaSE52M0U?oc=5Best in Grass Lands in New York, Inviting 2,600 Consumers to Judge the State’s Top Cannabis stupidDOPEFeb 17, 2026 8:00 AM
news.google.com
PARANORMASIGHT: The Mermaid's Curse (Switch) Reviewhttps://news.google.com/rss/articles/CBMimwFBVV95cUxQcWhoMGFLUG5ZYjNVVjV4ZFRiSlRCX2ZTdDAtME9qRTRVTl85RElSWXpuQUlFcVNGTlRreXNSV1FFcnc1ZVJrMnBHaGphU3VkZ1U3RFh5SFh4cDFxX0lHUEFUcWxEVURmWHlOWW0tUDJsSVRFSmdpZ0V3UXBWdEdoX2ZCalI2cFNhcnp0Z2h1aUJKUklIUmpyTHVvZw?oc=5PARANORMASIGHT: The Mermaid's Curse (Switch) Review Nintendo World ReportFeb 17, 2026 8:00 AM
news.ycombinator.com
Show HN: Decision OS – A framework to evaluate job offer risk (not just salary)https://news.ycombinator.com/item?id=47044706Hello HN,A few days ago I shared my "Should I Quit?" calculator. The most common feedback was: "Okay, I should quit, but how do I know the next thing isn't worse?"So I built Part 2: The Offer Evaluator.Most offer comparison tools just look at Total Comp or commute time. I wanted to model the structural risks of a career move using a weighted decision matrix.The Methodology: The "Decision OS" compares three states: Current Role vs. New Offer vs. The Market (Keep Searching). Instead of just looking at cash, it scores them on: Reversibility: If this role is a catastrophe, how hard is market re-entry? (High reversibility = lower risk). Sustainability: Calculated based on the gap between your current burnout levels and the new role's expected operational tempo. Downside Safety: Modeling the stability of the new company's runway vs. your current role. It produces a PDF "Dossier" that breaks down these variables so you can see the trade-offs clearly.I’m curious if you think "Reversibility" shFeb 17, 2026 7:37 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
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
github.com
Show HN: CodeGraph CLI – Chat with your codebase using graph-augmented RAGhttps://github.com/al1-nasir/codegraph-cliHey HN!I built CodeGraph CLI because I was tired of grep-ing through massive codebases trying to understand how things work.It combines three things: - tree-sitter (AST parsing, error-tolerant) - SQLite (dependency graph: nodes + edges) - LanceDB (vector embeddings, disk-based)The key insight: pure vector search misses structural relationships. So I combined vector search with BFS graph traversal — find semantically similar code, then expand to dependencies/dependents.Result: ask "how does authentication work?" and it finds validate_token(), its caller login_handler(), AND the dependency TokenStore — because it understands both meaning AND structure.Other features: - Impact analysis (multi-hop BFS: what breaks before you change it) - Multi-agent system via CrewAI (4 specialized agents) - Visual code explorer (browser-based) - Auto-generate docs/READMEs - 100% local-first (works with Ollama, zero data leaves machine) - 6 LLM providers (Ollama, OpenAI, Anthropic, Groq, Gemini, OpenRouterFeb 17, 2026 4:39 AM
bing.com
Why Your Hair Sheds More in Winter—According to Dermatologistshttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8cafda7000445d82d204a47ffcd5ae&url=https%3A%2F%2Fwww.cosmopolitan.com%2Fstyle-beauty%2Fbeauty%2Fa70382668%2Fwinter-hair-loss-explained%2F&c=18201596506238186340&mkt=en-usWe may earn commission from links on this page, but we only recommend products we love. Promise. Cold weather, lower humidity, and indoor heat can disrupt the hair growth cycle, pushing more strands ...Feb 17, 2026 4:38 AM
github.com
Show HN: M-Courtyard – Fine-tune LLMs on your Mac with zero codehttps://github.com/Mcourtyard/m-courtyardHi HN! I built M-Courtyard because fine-tuning LLMs locally was frustrating — CLI tools, Python scripts, and scattered documentation.This app wraps the full pipeline (document → dataset → fine-tune → test → export to Ollama) into a single desktop app that runs entirely on Apple Silicon.Tech: Tauri 2.x + React + mlx-lm + Ollama License: AGPL 3.0Would love feedback on what models or features would be most useful.Feb 17, 2026 3:52 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
news.ycombinator.com
Show HN: Provisioner per-board sidecar for serial access, flashing, and bring-uphttps://news.ycombinator.com/item?id=47040176Hi HN,I’ve been building a small infrastructure tool called Provisioner: a per-board provisioning sidecar for SBC bring-up and lab automation. https://github.com/alessandrocarminati/provisionerThe model is simple: one instance per board. It sits next to the hardware, owns the serial console, exposes access over SSH, and serves boot artifacts (kernel, initramfs, rootfs) via TFTP/HTTP. It’s designed for the phase where networking is unreliable, storage is half-flashed, and UART is the only thing still telling the truth.What it does- Serial console ownership + SSH multiplexing - Monitor shell + serial tunnels - Scriptable boot interaction - Artifact provisioning (TFTP + HTTP) - In-band file transfer over serial - gzip/base64 hacks + XMODEM - Derived from my earlier send_console-ng utility https://github.com/alessandrocarminati/send_file-ng - Pluggable power control - SNMP PDUs, among others cheap ESP8266/Tasmota devices - Serial logging - Lightweight access controlProvisioner also embeds Feb 16, 2026 8:52 PM
