\n From video to publish-ready AI subtitles\u2014all in one browser\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.856854"} +{"0": 2, "1": "producthunt", "2": "https://www.producthunt.com/products/z-ai", "3": "AutoClaw", "4": "AutoClaw.\n An AI work agent across desktop, browser, and chat\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.862656"} +{"0": 3, "1": "producthunt", "2": "https://www.producthunt.com/products/velofiler", "3": "VeloFiler", "4": "VeloFiler.\n Keyboard-first dual-pane file manager for macOS\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.867947"} +{"0": 4, "1": "producthunt", "2": "https://www.producthunt.com/products/pocket-by-meta", "3": "Pocket by Meta", "4": "Pocket by Meta.\n Vibe-code games, then share them like TikToks\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.872588"} +{"0": 5, "1": "producthunt", "2": "https://www.producthunt.com/products/port-radar-for-macos", "3": "Port Radar for macOS", "4": "Port Radar for macOS.\n An AI port manager for your Mac. \n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.877394"} +{"0": 6, "1": "producthunt", "2": "https://www.producthunt.com/products/pawvis", "3": "Pawvis", "4": "Pawvis.\n Control your mac through your webcam & train custom gestures\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.883214"} +{"0": 7, "1": "producthunt", "2": "https://www.producthunt.com/products/agents-never-sleep", "3": "Agents Never Sleep", "4": "Agents Never Sleep.\n Agents keep running with the lid closed\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.888204"} +{"0": 8, "1": "producthunt", "2": "https://www.producthunt.com/products/kerasformers", "3": "KerasFormers", "4": "KerasFormers.\n Keras 3 collection of pretrained models\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.892483"} +{"0": 9, "1": "producthunt", "2": "https://www.producthunt.com/products/maccess", "3": "Maccess", "4": "Maccess.\n Your Mac, in your pocket \u2014 trackpad, screen, and AI\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.899235"} +{"0": 10, "1": "producthunt", "2": "https://www.producthunt.com/products/zero-15", "3": "Zero", "4": "Zero.\n Vercel's programming language built for AI agents\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.904312"} +{"0": 11, "1": "producthunt", "2": "https://www.producthunt.com/products/open-analytics-2", "3": "Open Analytics", "4": "Open Analytics.\n AI-Native Google Analytics alternative for the modern web\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.908642"} +{"0": 12, "1": "producthunt", "2": "https://www.producthunt.com/products/toplify", "3": "Toplify", "4": "Toplify.\n Track your App Store ranking worldwide\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.913522"} +{"0": 13, "1": "producthunt", "2": "https://www.producthunt.com/products/actx0", "3": "Actx0", "4": "Actx0.\n Memory infrastructure for AI agents.\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.919382"} +{"0": 14, "1": "producthunt", "2": "https://www.producthunt.com/products/project-sky", "3": "Project SKY", "4": "Project SKY.\n Your ambient AI companion for Windows.\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.924724"} +{"0": 15, "1": "producthunt", "2": "https://www.producthunt.com/products/dockhand", "3": "Dockhand", "4": "Dockhand.\n Docker management for everyone\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.929012"} +{"0": 16, "1": "producthunt", "2": "https://www.producthunt.com/products/supernova-ai", "3": "Supernova", "4": "Supernova.\n All your data in Claude and Codex\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.933931"} +{"0": 17, "1": "producthunt", "2": "https://www.producthunt.com/products/wizstar", "3": "Wizstar", "4": "Wizstar.\n Digital avatars that move and act like professional actors\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.943295"} +{"0": 18, "1": "producthunt", "2": "https://www.producthunt.com/products/flunkey", "3": "Flunkey", "4": "Flunkey.\n Voice-first AI layer for Windows (beta)\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.949166"} +{"0": 19, "1": "producthunt", "2": "https://www.producthunt.com/products/epho-claude-code-in-the-cloud", "3": "Epho", "4": "Epho.\n Run Claude Code, Codex or Opencode in cloud with your repo\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.954553"} +{"0": 20, "1": "producthunt", "2": "https://www.producthunt.com/products/mindcase", "3": "Mindcase", "4": "Mindcase.\n Extract data from anywhere on the web within minutes\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.958994"} +{"0": 21, "1": "producthunt", "2": "https://www.producthunt.com/products/onecli", "3": "OneCLI", "4": "OneCLI.\n Give every employee a secured, sandboxed pro assistant agent\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.963846"} +{"0": 22, "1": "producthunt", "2": "https://www.producthunt.com/products/ramp-router", "3": "Router by Ramp", "4": "Router by Ramp.\n Tokens are money. Save both.\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.969872"} +{"0": 23, "1": "producthunt", "2": "https://www.producthunt.com/products/fx-by-vercel", "3": "fx (by Vercel)", "4": "fx (by Vercel).\n Vercel's tiny, open-source coding agent\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.975166"} +{"0": 24, "1": "producthunt", "2": "https://www.producthunt.com/products/google-antigravity", "3": "Antigravity IDE Extensions", "4": "Antigravity IDE Extensions.\n Antigravity agents now live inside your existing editor\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.981295"} +{"0": 25, "1": "producthunt", "2": "https://www.producthunt.com/products/shogunai", "3": "ShogunAI", "4": "ShogunAI.\n Your personal AGI on your PC. Built to finish real work.\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.987110"} +{"0": 26, "1": "producthunt", "2": "https://www.producthunt.com/products/pixelread-ai-ocr", "3": "PixelRead AI OCR", "4": "PixelRead AI OCR.\n Capture, translate, and understand any text on your Mac\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.992569"} +{"0": 27, "1": "producthunt", "2": "https://www.producthunt.com/products/local-7", "3": "Local", "4": "Local.\n Zero (!) friction local AI for your Mac\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:21.997011"} +{"0": 28, "1": "producthunt", "2": "https://www.producthunt.com/products/plow-latch", "3": "Plow Latch", "4": "Plow Latch.\n Run AI agents on your Mac with scoped access\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.004277"} +{"0": 29, "1": "producthunt", "2": "https://www.producthunt.com/products/lynqo-your-local-nas-server", "3": "Lynqo", "4": "Lynqo.\n Your machine is a P2P server, review suite & clipboard sync.\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.008899"} +{"0": 30, "1": "producthunt", "2": "https://www.producthunt.com/products/surfdeck", "3": "Surfdeck", "4": "Surfdeck.\n Your tabs, within reach.\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.016810"} +{"0": 31, "1": "producthunt", "2": "https://www.producthunt.com/products/outlook-google-calendar-sync-for-mac", "3": "Outlook Google Calendar Sync for Mac", "4": "Outlook Google Calendar Sync for Mac.\n Sync Outlook calendars to Google on your Mac\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.025053"} +{"0": 32, "1": "producthunt", "2": "https://www.producthunt.com/products/nobodywho", "3": "NobodyWho", "4": "NobodyWho.\n Run AI models on any device\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.029311"} +{"0": 33, "1": "producthunt", "2": "https://www.producthunt.com/products/shape-5", "3": "Shape", "4": "Shape.\n The agentic IDE for designers and programmers\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.038832"} +{"0": 34, "1": "producthunt", "2": "https://www.producthunt.com/products/lifelong-the-family-health-company", "3": "Lifelong", "4": "Lifelong.\n Your whole family\u2019s health in one place.\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.047781"} +{"0": 35, "1": "producthunt", "2": "https://www.producthunt.com/products/glasp-for-firefox", "3": "Glasp for Firefox", "4": "Glasp for Firefox.\n Highlight and summarize any page, PDF, or video in Firefox\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.056476"} +{"0": 36, "1": "producthunt", "2": "https://www.producthunt.com/products/meetstream-ai", "3": "MeetStream AI", "4": "MeetStream AI.\n Unified API & Infra for AI Meeting Agents\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.062820"} +{"0": 37, "1": "producthunt", "2": "https://www.producthunt.com/products/aloud-4", "3": "Aloud", "4": "Aloud.\n Turn spoken feedback into tasks your coding agent can run\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.070380"} +{"0": 38, "1": "producthunt", "2": "https://www.producthunt.com/products/berd", "3": "Berd", "4": "Berd.\n Weird, playful desktop app for building with AI agents\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.079903"} +{"0": 39, "1": "producthunt", "2": "https://www.producthunt.com/products/checksum-ai", "3": "Checksum AI", "4": "Checksum AI.\n Your coding agent\u2019s testing buddy\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.084304"} +{"0": 40, "1": "producthunt", "2": "https://www.producthunt.com/products/hynote-ai", "3": "HyNote for Mac", "4": "HyNote for Mac.\n Free local transcription that is 100% Private\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.089132"} +{"0": 41, "1": "producthunt", "2": "https://www.producthunt.com/products/grok-4-6-7", "3": "Grok 4.6", "4": "Grok 4.6.\n Frontier Intelligence for Long-Running Agents\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.095509"} +{"0": 42, "1": "producthunt", "2": "https://www.producthunt.com/products/hermai-brand-api", "3": "Hermai Brand API", "4": "Hermai Brand API.\n White label your B2B SaaS with every customer's brand\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.101518"} +{"0": 43, "1": "producthunt", "2": "https://www.producthunt.com/products/prized", "3": "Prized", "4": "Prized.\n Let non-engineers build secure internal tools \n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.113643"} +{"0": 44, "1": "producthunt", "2": "https://www.producthunt.com/products/calendly", "3": "The New Calendly", "4": "The New Calendly.\n Handle all of the work before, during, and after meetings\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.125681"} +{"0": 45, "1": "producthunt", "2": "https://www.producthunt.com/products/minimax", "3": "MiniMax Design", "4": "MiniMax Design.\n Your own agent team for open-ended creation\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.132779"} +{"0": 46, "1": "producthunt", "2": "https://www.producthunt.com/products/zoho", "3": "Zoho Cliq 7.0", "4": "Zoho Cliq 7.0.\n Uninterrupted work\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.139594"} +{"0": 47, "1": "producthunt", "2": "https://www.producthunt.com/products/peach-co-pilot", "3": "Peach Co-Pilot", "4": "Peach Co-Pilot.\n WhatsApp Sidekick for busy professionals\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.150048"} +{"0": 48, "1": "producthunt", "2": "https://www.producthunt.com/products/revy-the-fashion-app", "3": "Revy", "4": "Revy.\n The ownership layer for fashion, shopping, and your wardrobe\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.158777"} +{"0": 49, "1": "producthunt", "2": "https://www.producthunt.com/products/cloudways", "3": "Cloudways Managed AI Agents", "4": "Cloudways Managed AI Agents.\n Skip the setup and run OpenClaw & Hermes, fully managed\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.167625"} +{"0": 50, "1": "producthunt", "2": "https://www.producthunt.com/products/roveri", "3": "Roveri", "4": "Roveri.\n A riding journal for iPhone every ride, painted on a map\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-22T18:05:22.181600"} +{"0": 51, "1": "hackernews", "2": "https://ringlochid.me/imagesage/index.html", "3": "Show HN: Find and Organize Photos with Private, Local AI", "4": "Show HN: Find and Organize Photos with Private, Local AI. ", "5": "2026-08-22T18:05:23.068371"} +{"0": 52, "1": "hackernews", "2": "https://github.com/dat999zx/knowl", "3": "Show HN: Knowl \u2013 CLAUDE.md hit 1000 lines, so I built memory that prunes itself", "4": "Show HN: Knowl \u2013 CLAUDE.md hit 1000 lines, so I built memory that prunes itself. Using AI agents is amazing, but in long-term, I started to have more and more problems about the context of them. They keep forgetting what we are working on, opening a new chat session clears all their context, my CLAUDE.md had like 1000 lines.So I thought to install those agent memory that are on the market. Started out great, but I noticed that those memory only append and does not fix what is stale. First day I told it to use Lemon Squeezy as our MoR but second day I tell it to change to Polar. But when I ask it again, they keeps returning both answers or cannot decide which we are using.
That's why I created Knowl. Knowl solves the problem with freshness of the knowledge. It split the knowledge into small data bits called "atom". Atoms can be of the following types: fact, decision, goal, constraint, architecture, state, skill so we can retrive atoms in categories.
When conflict happens at write-time (a new atom conflicts with an old one), Knowl retires the old one (flag it with superseded) and remove it out of main retrieval but still keeps full history.
There are many more cool features like transcript search, multi-workspace sharing, change detection impact... and Knowl Cloud for team-sync too.
We benchmarked Knowl on MemoryAgentBench - FactConsolidation single-hop @262K context and got suprising result:
- Knowl: 0.90 <- I ran this
- agentmemory: 0.79 <- and this
- Gpt-4o (full context): 0.60
- Mem0: 0.18
- Zep: 0.07
In multi-hop we scored 0.07 (ceiling of all time is 0.14)\nI ran at temperature 0.7. You can find full benchmark in my repo.
This is fully open-source, connect to almost every providers through MCP (Claude Code, Codex, Cursor, Antigravity...). And it is fully local (unless you use Knowl Cloud).
I'd love to get some feedback. Cheers!", "5": "2026-08-22T18:05:23.075097"} +{"0": 53, "1": "hackernews", "2": "https://github.com/kidus-tiliksew/conveyor", "3": "Show HN: I make my AI agents file paperwork before they're allowed to code", "4": "Show HN: I make my AI agents file paperwork before they're allowed to code. ", "5": "2026-08-22T18:05:23.081569"} +{"0": 54, "1": "hackernews", "2": "https://github.com/pawaca/dsh-edge", "3": "Show HN: Running a full AI coding agent inside Cloudflare Durable Object", "4": "Show HN: Running a full AI coding agent inside Cloudflare Durable Object. ", "5": "2026-08-22T18:05:23.086769"} +{"0": 55, "1": "hackernews", "2": "https://knoku.com", "3": "Show HN: Knoku \u2013 cited AI answers from docs, files, and team knowledge", "4": "Show HN: Knoku \u2013 cited AI answers from docs, files, and team knowledge. ", "5": "2026-08-22T18:05:23.091010"} +{"0": 56, "1": "hackernews", "2": "https://vinv.ai/", "3": "Show HN: VinvAI runs and observes your services to find bugs and perf issues", "4": "Show HN: VinvAI runs and observes your services to find bugs and perf issues. ", "5": "2026-08-22T18:05:23.095775"} +{"0": 57, "1": "hackernews", "2": "https://valeriavg.dev/agile-ai-development-lifecycle", "3": "Show HN: Agile AI Development Lifecycle", "4": "Show HN: Agile AI Development Lifecycle. Schr\u00f6dinger's AI-DLC Framework", "5": "2026-08-22T18:05:23.101546"} +{"0": 58, "1": "hackernews", "2": "https://learnleap.xyz", "3": "Show HN: Learn Leap, an AI tutor that teaches from your own material", "4": "Show HN: Learn Leap, an AI tutor that teaches from your own material. I built Learn Leap because I found myself constantly asking ChatGPT questions while reading research papers. I wanted something that already understood what I was reading. With Learn Leap you can upload your own material and have an AI tutor teach directly from it. The tutor has full context on your material and can create quizzes, flashcards, and exams to test your understanding.\nIt's still early and I'd love feedback from HN on the product and the learning experience. Cheers.", "5": "2026-08-22T18:05:23.106771"} +{"0": 59, "1": "hackernews", "2": "https://krishna-modi12.github.io/frontend-design-pro/", "3": "Show HN: Front end skill pack for AI agents, with machine-enforced quality gates", "4": "Show HN: Front end skill pack for AI agents, with machine-enforced quality gates. ", "5": "2026-08-22T18:05:23.111026"} +{"0": 60, "1": "hackernews", "2": "https://www.anjadhe.com/demo", "3": "Show HN: Anjadhe \u2013 privacy first AI assistant, no account, no server DB", "4": "Show HN: Anjadhe \u2013 privacy first AI assistant, no account, no server DB. Hi HN, I am Ram. For the last few months I have been building Anjadhe. It is a personal AI assistant for macOS. The main idea is simple: A macOS app where AI does the job of a personal assistant for the user. Not a chat only app where user needs to dig through chats to understand today\u2019s schedule or a project they planned with the ai last week, but a canvas where the user and AI work together around basic tools.
I started building this as DYI tool for myself, but at this point I see that its truly useful for me, so sharing with others to see if it can be useful to others too.
Here is a quick Demo - https://www.anjadhe.com/demo
What it can do today:
- It reads your incoming email and files the important things: bills with due dates, renewals, receipts, deliveries. A bill becomes a task with a date, by itself. Two booking emails become one trip.\n- You set goals by talking to it, not by filling forms. It asks you questions, makes a plan with dated tasks. Later you can say "work got crazy, push everything two weeks" and it moves the whole plan. It always shows you what will change and asks before doing it.\n- Routines run without you. On a schedule, or when a certain email or file arrives. Every run leaves a log you can read.\n- You can give it some documents you wrote. It learns your writing style and writes new content in your voice. What it learned is a page you can read and edit, not a black box.
About privacy: the AI runs on your Mac with llama.cpp. Or you can point it to your own server, or use your own OpenAI/Anthropic key, or use Anjadhe Cloud (open-weight models, free allowance, no account needed). Your data stays in SQLite on your own disk. No account. No telemetry unless you turn it on. The source code of the app and the cloud service is public, so you do not have to trust my words.
Honest limitations: it is an early alpha, macOS only. Local models need a Mac with atleast 32gb ram to work. The email features work fine on small models, but the full agent wants a bigger model or a server. Also it is Electron with vanilla JS, no framework. Looking forward to hear your opinions.
Demo (2 min): https://anjadhe.ai/demo\nDownload: https://anjadhe.ai/download\nSource: https://github.com/Anjadhe/Anjadhe
I would love feedback, especially if you think my privacy claims are wrong somewhere. I will be here to answer questions.", "5": "2026-08-22T18:05:23.117708"} +{"0": 61, "1": "hackernews", "2": "https://github.com/ArihantDeva/heimdall", "3": "Show HN: Heimdall \u2013 Trust-verified knowledge layer for AI coding agents", "4": "Show HN: Heimdall \u2013 Trust-verified knowledge layer for AI coding agents. ", "5": "2026-08-22T18:05:23.122782"} +{"0": 62, "1": "hackernews", "2": "https://app.flightledger.net", "3": "Show HN: Flight Ledger \u2013 Track your flights, what they cost, and United status", "4": "Show HN: Flight Ledger \u2013 Track your flights, what they cost, and United status. Hello HN,
Flight Ledger is a private ledger of your flights, focused on (but not limited to) the United Airlines ecosystem. I am a physicist who has been based near two United hubs, first SFO and now IAH, and I fly United a lot, mostly for work.
I built this app to track my flights, flight expenses, and Premier status qualification without having to log into various accounts. All the imports (such as MileagePlus activity csv, .eml receipts, or Flighty/myFlightRadar24 exports) and data filling are manual.
A primary feature is that the app requires no sign-up and has no backend. Everything is stored locally, using SQLite and WebAssembly. This means that frequent back-ups are advisable to avoid losing data. For more robust backups and multi-device sync, the app also supports sync via Google Drive. I use this option for myself, but would be curious about other possible solutions.
Perhaps the main features that differentiate it from other apps are that the app tracks costs and metrics like CPM (cost per mile), United-specific PQP, PQF, and award miles, and allows one to differentiate out-of-pocket and reimbursed expenses. It also includes other things such as route tracking and a lifetime mile tracker.
While I have extensive programming experience in scientific computing and C++, my experience in web applications is more limited, and this app was built with Claude Code, which is unsurprising in this day and age, I guess. It is open source: https://github.com/vlvovch/flight-ledger
Curious to see your thoughts. I have particularly struggled with parsing various receipts, especially ticket reissue chains, and will be interested to know if the parser will survive yours.", "5": "2026-08-22T18:05:23.126997"} +{"0": 63, "1": "hackernews", "2": "https://benhoyt.com/writings/updating-gifty-with-ai/", "3": "Show HN: Updating my wedding registry website with AI in 275 commits", "4": "Show HN: Updating my wedding registry website with AI in 275 commits. ", "5": "2026-08-22T18:05:23.133753"} +{"0": 64, "1": "hackernews", "2": "https://paloaltocivic.com", "3": "Show HN: AI driven civic dashboard for Palo Alto", "4": "Show HN: AI driven civic dashboard for Palo Alto. ", "5": "2026-08-22T18:05:23.138788"} +{"0": 65, "1": "hackernews", "2": "https://kinodesk.net/", "3": "Show HN: Kinodesk \u2013 Unlimited Remote Desktop Sharing", "4": "Show HN: Kinodesk \u2013 Unlimited Remote Desktop Sharing. Kinodesk is a new advanced remote desktop sharing software that aims to reach the full performance by taking advantage of advanced performance enhancing utilities to reach 2K 240fps on capable hardware to ensure you have a smooth experience.", "5": "2026-08-22T18:05:23.143009"} +{"0": 66, "1": "hackernews", "2": "https://www.chickenbutt.dev/", "3": "Show HN: ChickenButt a Native GTK Chat Client for Ollama on Linux", "4": "Show HN: ChickenButt a Native GTK Chat Client for Ollama on Linux. What's up, ChickenButt?
I made a free, native GTK client called ChickenButt. :)
It lets you chat with local models through Ollama, and I figured some of you might get a kick out of it.
Here's the repo: https://github.com/pixelhackstudios/ChickenButt
I've always hated reading long text in the terminal, and I couldn't find anything that ran locally that allowed me to get quick answers from LLMs when I needed them.
Also, not everyone has access to the frontier models and whatnot, so I thought helping people who only have access to more cost-effective AI would be a fun way to contribute to the FOSS community. So, I built ChickenButt!
It was really fun to build, and I wanted to share it.
Enjoy! :)", "5": "2026-08-22T18:05:23.147818"} +{"0": 67, "1": "hackernews", "2": "https://ozbrain.com", "3": "Show HN: OzBrain, a shared brain for knowledge between agents and your team", "4": "Show HN: OzBrain, a shared brain for knowledge between agents and your team. I think agent-first chat interfaces will be a primary software modality and busy dashboard/UI will go away. I\u2019m not sure who exactly wins it, but I want my knowledge to grow/go with me.
A lot of the \u201cknowledge\u201d ie research, analysis, reasoning will be done by agents as the primary user. Our current notes tools & tasks management systems were built for humans\u2026 I don\u2019t care what the 17th thing on my bug backlog is. I want to conduct agents that can execute for me and do great work.
What I built OzBrain to do:\n+ Create a central place for agent reasoned knowledge to live\n+ Be agnostic about what apps/agents connect to it\n+ Capture everything and track it so I can audit it\n+ Enable teams, collaborators or partners to share brains\n+ Handle conflicts so many agents in the same article doesn\u2019t blow up\n+ Refactor knowledge into more token friendly chunks and map the index well\n+ Close the knowledge loop so new thinking supersedes old thinking across the corpus. Don\u2019t erase, depreciate and link\n+ Keep user data safe and secure\n++ Be easy enough to use that you don\u2019t have to have any technical knowledge
Some among us will always build their own custom solutions, but there are millions of tech professionals and small business owners that will use agents heavily and need a solution. So I\u2019m trying to build that.
Isn\u2019t this like gBrain? Yes, similar. I think it\u2019s like AWS vs Vercel. AWS is very powerful, configurable, and useful if you\u2019re technical and want to invest the time into really fine tuning your system\u2026 but if you just want your web deploy/hosting to just work and be easy to deal with you use Vercel.
// WHY I MADE IT
I\u2019ve been enjoying getting back to my technical roots, as I lost my coding skills more than a decade ago, but with AI I can focus on the system and the product in partnership with agent coding workflows.
I recently built a Voice AI for older people. To build it I created an agentic engineering workflow (feel free to rip that up as I\u2019m always looking to improve systems: https://ozbrain.com/resources/eng-flow) My approach with coding agents is trust but verify, and I\u2019m trying to replace the parts where a human would review with an adversarial or specialized agent who would give a better answer/review.
I have workflows that will go high level task to shipped PR running in Claude cloud sessions. I use Claude Code locally and Cursor when I want a tighter loop on doing visual work like UI or layout. And Codex to either load balance usage for TokenThriffting or when I want a different llm to think thru something.
It was a pain in the ass passing .md files around and keep track of which version was the most recent, so I built a hosted .md storage right in Supabase and any of my agents already have Supabase access. This let me build a solid, scalable, secure voice AI from my phone at the gym. All my agents have access to our knowledge, can write to it, update and refer to it as we build and improve the product and the systems we use.
Out of 75 founder friends I asked about how they manage shared knowledge, 26 built their own custom knowledge systems\u2026 Obsidian vaults with 7k files synced through a VPS, markdown repos behind their own MCP servers, cron jobs stitching Supabase to a skills file\u2026 each a different Frankenstein they have to maintain. 32 said they felt the pain of moving static files around but didn\u2019t have any solution for it.
So I rebuilt my brain better and used it to build it.
// HOW YOU CAN HELP
Would love to have you try it out. The maintenance loop is still in alpha so not running it on customer data yet.
If you built your own brain I\u2019d love to hear how you did it. What criteria was most important for you in its design & function.
If you are tired of shuffling .md files around I\u2019d love to have you try out OzBrain and to give feedback, just ask your agent to put it in the shared bugs & features brain!
Cheers!\nBubs.co", "5": "2026-08-22T18:05:23.153575"} +{"0": 68, "1": "hackernews", "2": "https://frontpageoftheinternet.lol/", "3": "Show HN: Front Page of the Internet", "4": "Show HN: Front Page of the Internet. you can buy the front page of the internet and stay there, while you use it for AI SEO, until someone out bids you and now they are on the front page of the internet, bid now just for a dollar", "5": "2026-08-22T18:05:23.160135"} +{"0": 69, "1": "hackernews", "2": "https://github.com/manishrjain/zroar", "3": "Show HN: Zroar \u2013 Serialized Roaring Bitmaps in Zig", "4": "Show HN: Zroar \u2013 Serialized Roaring Bitmaps in Zig. zroar is a ground-up implementation of Roaring Bitmaps data structure in Zig. zroar stores both the keys and (array, bitmap) containers in a single flat byte buffer, making the in-memory representation equal to the on-disk or over-the-network representation, eliminating the serialization/deserialization step entirely.
The design was originally aimed at systems which keep their posting lists on disk, but zroar performs faster than CRoaring even for purely in-memory ops, due to CPU cache locality.
Against CRoaring 5.0's benchmarking suite (ported to Zig), zroar is faster in 339 out of 360 tests, being 2x-9x faster (geometric mean), and up to 600x faster on serialize/deserialize.
zroar avoids complex mechanism (like adaptive radix trees), uses Zig native SIMD ops and is simpler. The main logic is written in ~2000 lines of code, while CRoaring's 64-bit bitmap codebase is over 17000 LOC.
Not yet: By choice, zroar doesn't support run containers, and is 64-bit only.
There are more details in the GitHub README. Try it out! I'd love feedback on the API and design. zroar is a Zig-based successor to my other project, sroar in Go, which showed a similar boost. So, I think this design should show performance gains in any language.", "5": "2026-08-22T18:05:23.164917"} +{"0": 70, "1": "hackernews", "2": "https://traccia.ai/", "3": "Show HN: Traccia - Observability, Runtime Control & Audit for agents", "4": "Show HN: Traccia - Observability, Runtime Control & Audit for agents. AI applications are becoming agents, which has started to take autonomous decisions. There are plenty of tools and platform available to trace, and observe what an agent or llms calls does. They are good in what they do, but tracing and observability isnt enough for AI agents era. We need a solution that can help you observe, evaluate, create run time policies to govern and finally audit the actions of the agent. We built Traccia to solve this problem. The good part, all of these can be achieved by just writing few lines of code. Traccia has an open-sourced sdk that can work with your existing observability tool like grafana, tempo, jaeger, etc. In case you need more than just observability, Traccia provides the platform to evaluate, control and audit the agents. The platform is easy to use. The product's documentation is quite extensive. It is also cloud vendor and framework agnostic. Traccia is being built by an Indian start up ,based out of Bengaluru. We are running a 3 months free trials so that you can explore without any strings attached. We are open to improvise and get better so please drop your comments and feedbacks.", "5": "2026-08-22T18:05:23.170811"} +{"0": 71, "1": "hackernews", "2": "https://aristralabs.com", "3": "Show HN: Aristra, your life OS", "4": "Show HN: Aristra, your life OS. Most AI cares only about the outcome users want. Aristra cares about your process. It works alongside and for you across the things you already use, learning about you and from you, day and night. What works becomes a playbook it reaches for next time. What stops being true quietly fades. How you actually get things done is the most valuable thing about you, and it's the one thing no AI understands. One AI. One memory. Your life OS.\nLaunch vid: https://youtu.be/CoIKF_CKE8g?si=v5QeyHxJvITrCNAy", "5": "2026-08-22T18:05:24.522259"} +{"0": 72, "1": "hackernews", "2": "https://www.basecompute.co/local", "3": "Show HN: Zero () friction local AI for Mac", "4": "Show HN: Zero () friction local AI for Mac. Super excited to launch our new app Local today. What we\u2019ve learned at Base Compute over the last months is that running AI directly on your laptop or workstation gives you maximum privacy and it\u2019s free, but it\u2019s also a massive headache to configure. So we\u2019ve decided what matters is making the experience completely frictionless for users.
Local analyses the hardware of your laptop, optimises the AI for it, and recommends the best models for your specific device.
It let\u2019s you do what you\u2019re doing with cloud AI already, just for free and on your own machine: Chatting with PDF\u2019s, Recording and summarising meetings, running coding agents...
If you\u2019re using Local in your office with colleagues, you can run it in \u201cOffice Mode\u201d. The strongest computer in your office runs the AI and everyone can connect to it with their laptop. The data never leaves the office.
It\u2019s available for download on our website today, please try it out and let us know what you think!", "5": "2026-08-22T18:05:24.534254"} +{"0": 73, "1": "hackernews", "2": "https://github.com/runvendo/vendo", "3": "Launch HN: Vendo (YC S26) \u2013 Let users build features on top of your product", "4": "Launch HN: Vendo (YC S26) \u2013 Let users build features on top of your product. Hi HN, we\u2019re Yousef & Nour, founders of Vendo (https://vendo.run). Vendo lets users create new features inside the software they already use. A user describes the dashboard, workflow, or small app they need, and Vendo builds it on top of the product\u2019s existing data, API, and interface.
Demo: https://www.youtube.com/watch?v=VdpHehY64ls
We built Vendo because every SaaS eventually faces the same problem: every customer needs something slightly different. One wants a new report and another needs a workflow that only makes sense for their team. These requests either sit on the roadmap, become one-off engineering work, or force the customer into spreadsheets and external tools. We wanted the user to be able to create the missing feature themselves, without leaving the product.
Here is how it works:
- npx vendo init reads the product's API surface, theme, routes, and more. These are used so that the apps Vendo creates (1) look on-brand and native and (2) have the ability to read data and perform actions directly through the company's API
- When a user asks for a feature, we have a custom Vendo harness that writes a React component with a bunch of Vendo add-ons and guardrails (ex. ability to make calls to the host API + our component library). Every save is compiled, type-checked, run against real API responses, and rendered before the user sees it. We just released a benchmark and write-up here with more info for anyone interested: https://vendo.run/blog/generating-product-ui-measured
- We use QuickJS to make sure that anything the agent creates is sandboxed and can't mess with the company's site. Vendo compiles the component and runs it with Preact inside a QuickJS VM with no access to the DOM, network, or clock. The VM returns a UI tree, which the host renders using the product\u2019s registered components. When the user clicks something, QuickJS emits a tool call; the host executes it through Vendo\u2019s guard and passes the result back into the same VM, preserving the screen\u2019s local state.
There's a lot of generative UI right now: streaming developer-written components into a chat (Vercel AI SDK, CopilotKit, Thesys), or rendering your app inside someone else's assistant (OpenAI Apps SDK, MCP Apps). We differ on two things. Vendo lives in your product and acts through your API as the signed-in user, so what it makes is durable: real apps users keep, pin, and run on triggers while they're away, and not components that are merely confined to a chat. Plus, it's not capped at putting together a bunch of prebuilt components: the agent can build arbitrary apps, from a quick dashboard out of your own components to real custom code running in a sandbox, and either way data only ever comes from tool calls to your API.
Here are some things customers are using Vendo for today:
- Letting their users create custom dashboards and reports. These are mainly UI-based and focused on letting the user see the exact graphs and metrics they care about
- Letting their customers create recurring automations. A big thing as well that has been used for these automations is the fact that we connect to external connections, so users have been automating many of their inter-tool workflows (ex. an automation that sends a slack alert based off of something in the product)
- B2B customers letting their customers customize the product with specific business logic. Often this is simple things like an extra field on a form, or an extra permission, but it is hard for a business to keep up with them otherwise.
- Creating and sharing custom dashboards/apps across an organization. Since the apps Vendo creates are durable, they can be shared, reused, and forked (which can\u2019t be done with many of the other in-chat generative UI solutions)
We've spent a lot of time thinking about how AI and agents will change the way people consume software. We think the answer is personal(ized) software: you see the UI you need to see, you tell an agent exactly what you need, and the product molds to how you work.
The key insights that have enabled the product to work are:
- A rule in code always beats a rule in a prompt.
- Invent as little syntax as possible. Generation got faster and more reliable when the output looked like what models already know (JSX-shaped markup) instead of a clever custom format.
- Deterministic beats model wherever you can get away with it. Theme extraction is pure static analysis, and a remix starts as a copy of your component, no model call.
Vendo is completely open-source (Apache-2.0) and can be self-hosted, so feel free to c", "5": "2026-08-22T18:05:24.539005"} +{"0": 74, "1": "hackernews", "2": "https://aiomniu.top/services/resume-screening", "3": "Show HN: AI Resume Screening \u2013 Looking for feedback from recruiters", "4": "Show HN: AI Resume Screening \u2013 Looking for feedback from recruiters. In addition to the resume creation tool we mentioned in our previous post, we've also developed an AI-powered resume screening tool to complement our overall ecosystem. This tool aims to help recruitment teams efficiently review hundreds of resumes without relying on keyword matching.
This tool helps teams:
- Batch upload and parse PDF resumes
- Compare candidates with job descriptions
- Review candidates' skills, experience, project experience, and industry fit
- Provide candidate fit analysis and scoring ranges
- Export results as a CSV file
To better meet the requirements of the European and American markets, candidate contact information is encrypted by default when uploading data. We do not provide judgments on unsuitable or unrecommended candidates. Artificial intelligence can only be used as an aid, not a replacement for HR judgment. We oppose all forms of professional discrimination.
We are currently offering Beta testing access to 100 HR professionals, recruiters, hiring managers, and technology team leaders. Beta users can enjoy:
- Free use during the testing period
- 1800 points per month after official launch
- 35% discount on on-premises deployment services
If you work in recruitment, you are welcome to participate in the trial:
I would like to understand the issues with false positives, false negatives, and shortcomings in the workflow. What is the biggest pain point in your current resume screening process?", "5": "2026-08-22T18:05:24.548030"} +{"0": 75, "1": "hackernews", "2": "https://news.ycombinator.com/item?id=49367763", "3": "Ask HN: We have company runway but not founder time. When do you go full-time?", "4": "Ask HN: We have company runway but not founder time. When do you go full-time?. Three first-time founders in Canada, all of us between 45 and 55: business/product, marketing, and me on the technical side. We're building an AI tool for creative work. About thirty testers have been on a pre-production build for a couple of months. Feedback is good, and importantly is not from friends, but actual target users who we sourced through our networks, which I'm treating as encouraging signal on product-market fit.
We're bootstrapped with CAD $25k initial commit and an additional up to $100k available, though we'd rather earn the right to spend each additional dollar than commit it up front. All three of us still have day jobs, cleared with our employers. At current near-zero usage we could run the service for ~18 months without revenue \u2014 that number obviously falls as real users arrive, which is part of the problem.
So our constraint isn't the usual one. The company has runway; founder time doesn't. I built the platform, I'm the one who has to go full-time first, and I'm also the highest-paid person on the team with family obligations that rule out a sudden income cut.
For those who started companies in their 40s or 50s: when did you know it was time to leave the day job? Did you wait for revenue to cover you, cut to part-time, build personal runway first, or raise specifically because founder availability had become the bottleneck?
Not naming the product \u2014 I am looking for advice, not a launch thread.", "5": "2026-08-22T18:05:24.555312"} +{"0": 76, "1": "hackernews", "2": "https://github.com/onecli/onecli", "3": "Launch HN: OneCLI (YC S26) \u2013 OSS sandboxed agent harness for teams", "4": "Launch HN: OneCLI (YC S26) \u2013 OSS sandboxed agent harness for teams. Hi HN, Jonathan & Guy here from OneCLI, an agent harness built for teams, giving every employee a secured, sandboxed personal agent.
Here\u2019s what you can do with it:
1. get a sandboxed agent, with all the OneCLI capabilities in place like connect your GitHub account, Gmail, Notion, or Dropbox simply from the chat.
2. deterministic human in the loop approval in the chat itself for things that you need 100% control like sending an email or deleting the Linear ticket.
3. manage team policy in one place, enforced across every agent in the workspace
4. enjoy global connections at the team level, like shared LLM keys or service accounts
Here\u2019s a demo: https://www.youtube.com/watch?v=dlW-44ntpbE
We started working on this by accident, even though our careers were in the security space. We were working on a devtool called ChartDB, an open-source DB tool. When OpenClaw took off back in January, we started using it to orchestrate agents on top of ChartDB. We quickly understood there is a big issue around auth. Agents need credentials to do real work, but to give them those secrets would not be the best idea. They keep them in their memory and also write them down to local files and their sessions as plain text. And we knew that agents can easily be fooled into giving up those API keys/secrets. So we needed some way to control the agent and stop prompt injections from tricking it into using its services for an attacker's benefit.
We created OneCLI that started as a vault for AI Agents built in Rust.
We found out that most of our demand for OneCLI came from autonomous agents like Hermes, OpenClaw and NanoClaw for individuals and teams.
Users looked for useful agents that do things for the person who runs them with two missing parts: 1) managing secrets and permissions. 2) and for teams - multiplayer management.
We decided to pivot and provide the agent itself as a harness for teams, to give each employee an agent. We saw that teams had to deal with setting up their own harness again and again, and basically as we already had the vault as a gateway. We got the idea to provide the missing piece of the agent management out of the box and open source it (Apache-2.0, with a small enterprise exception).
We're open source first - the entire platform, not just a small portion of it like other agents, so companies can actually see the code, evaluate it, and trust it instead of taking our word for it. They run it isolated, in their own environment, fully under their control, at production quality, not a locked black box hosted somewhere else. That means the safety isn't just a promise, it's something they can verify themselves. Combined with real autonomy and least-privilege access, that's what makes it something a company can fully own and trust, not just adopt.
We also approach this from a company perspective rather than an individual one. Our solution manages agents on behalf of each employee, wrapped in deterministic guardrails that company admins configure through centralized policies.
For the agent engine itself we\u2019re using jcode which is the core of the agent-loop. We found out that it improves the experience and makes the agent smarter and faster.
Here\u2019s how it works:
It runs on infra you control. Fully open-source, self-host or cloud in minutes.
The agent never holds a real secret. It gets a placeholder. The real credential is injected at the gateway, per request, after the call is authorized. It never enters the agent's context, memory, or logs.
Enforcement outside the model. Prompts are suggestions. Policies defined by the org admin run at the network layer, outside the agent and the LLM. Block endpoints, rate limit per agent, require approval, scope per employee. The gateway decides. The agent can't bypass it.
Isolated VM per agent. Own memory, own keys, own permissions. Blast radius is one agent.
Speed of the Harness: Rust engine under the agent loop.
Full identity trail. Every agent is bound to an employee. Every call logged with who it acted for and which policy allowed it.
Some things people are doing with the platform include:
- Managing their company life cycle entirely from the sales calls, to the product side automatically open tickets to the engineering teams, that would kick the development agents to deliver and ship to production.
- Operational side, like automatically hygiene the CRM after calls, sourcing leads, book meetings and manage follow ups emails.
- Some of our customers also doing their entire grocery shopping using those agents and send them to take care of their chores like ordering things online.
About the team: Both founders come from cybersecurity backgrounds. Jonathan spent years at Axis Security building zero trust network acces", "5": "2026-08-22T18:05:24.562300"} +{"0": 77, "1": "hackernews", "2": "https://deftwriting.com", "3": "Show HN: Deft Writing, an AI lab for non-slop LLMs", "4": "Show HN: Deft Writing, an AI lab for non-slop LLMs. Announcing Deft, a new AI lab for better writing
Currently, 86% of user queries are fully human according to pangram.
This is still a small beta model and it might make mistakes. We are launching our public beta now to get more feedback before scaling up.
Tips for better performance: \n- Add more details to your prompt. If you just provide a short sentence prompt, it will likely get detected as AI.\n- Try changing the style in "advanced options"\n- Deft currently works better for some use cases like Analysis/Essays, Creative writing, and Rewrites. It works less well for Marketing copy and news articles.
Our main goal is better writing, fooling AI detectors is just a side effect.
Here's the launch announcement the Deft model wrote for itself:
Introducing Deft Writing\nWe are proud to announce the launch of our new startup, Deft Writing. Having spent many years in the trenches with other Large Language Models (LLMs), we were continually hampered by poor writing. We believe writing well is hard. So we're making it easier for you to share your ideas, and by doing so make those ideas more readable for those who have to read them. Deft Writing saves you time writing and allows you to focus on your ideas.\nWe've developed a new training algorithm for models called Distribution Fine Tuning (DFT) that makes model outputs more human-like. Many other LLMs produce outputs riddled with overused "slop" phrases. DFT discourages this bad behavior.\nOur current offerings:\n* Our beta model is available for you to try out.\n* We also provide API access to the model.\n* For larger enterprises that want a model in their own unique style, we can train a custom model on your data to your specifications.\nGo check us out at www.deftwriting.com and try it out.", "5": "2026-08-22T18:05:24.568764"} +{"0": 78, "1": "hackernews", "2": "https://aiomniu.top/services/resume-builder", "3": "Show HN: AI Resume Optimization Tool for Specific Job Positions", "4": "Show HN: AI Resume Optimization Tool for Specific Job Positions. Hi HN, I've been watching everyone's presentations as an observer, and today I can finally showcase my product!
It helps users:
- Organize roles, projects, skills, and achievements
- Transform vague descriptions into clearer, more professional language
- Adjust the focus according to different target positions
- Check resumes before submission to the ATS system
Its core principle is: AI should enhance expressiveness, not fabricate experience or create false achievements.
Relatively speaking, this isn't my proudest achievement, as resume optimization isn't a new tool, but it's part of my main project's ecosystem.
Regarding the models used, our site's AI tool primarily uses Deepseek v4 flash and glm5.2. We will upgrade to the latest versions as the overall model evolves, such as the current glm5.3. We will be offering Beta testing access to 100 job seekers. Beta testers will receive:
- Free use during the testing period
- 900 credits per month after launch
- A free resume webpage after launch
Students, career changers, and experienced professionals are welcome to try it:
I especially welcome feedback on the accuracy and usefulness of the suggestions and what shortcomings the tool currently has. What is the most frustrating part of writing or improving a resume?", "5": "2026-08-22T18:05:24.578143"} +{"0": 79, "1": "hackernews", "2": "https://slaunt.ai", "3": "Show HN: Slaunt \u2013 control what AI agents can access, do, and execute", "4": "Show HN: Slaunt \u2013 control what AI agents can access, do, and execute. Hey HN, wanted to silently launch what I've building for a bit. Lmk.", "5": "2026-08-22T18:05:24.587728"} +{"0": 80, "1": "hackernews", "2": "https://tesana.ai/en", "3": "Show HN: Tesana \u2013 An AI game engine that builds quality games end-to-end", "4": "Show HN: Tesana \u2013 An AI game engine that builds quality games end-to-end. Hey HN! I built https://tesana.ai/en, a game generation platform that allows anyone to create full games with graphics, game logic, and animations from a simple text prompt - without coding.
Tesana has three key features\n1. describe your game idea and have Tesana write a game plan for you \n2. Build the game iteratively with AI, or let AI self-build the game in a build LOOP.\n3) Launch the game on the web, without setting up an engine
Tesana has 250 000 builders on the platform today
I believe the democratization of game making is a good thing, and hopefully Tesana can help more people bring their game ideas to life.
Happy to chat about the AI or dev stack behind the product. Let me know if you have any\nquestions/comments/feature requests!", "5": "2026-08-22T18:05:24.592931"} +{"0": 81, "1": "hackernews", "2": "https://openwebsearch.ai", "3": "Show HN: OpenWebSearch \u2013 A router for web search indexes", "4": "Show HN: OpenWebSearch \u2013 A router for web search indexes. Hi HN, I'm one of the people behind OpenWebSearch (https://openwebsearch.ai).
It's a router for web search indexes. You POST to one endpoint with a\n`provider` field, and it normalizes both the request and the response for web search indexes like Parallel, Brave, Exa and more
Why we built it: we run a model company (Interfaze) and a lot of our models are smaller in size and we're experimenting if given web search can a smaller 9b or 70b model perform the same as 300b or 600b model and we found that it does extremely better when given web search similar to this paper (https://arxiv.org/abs/2203.05115)
but we also found not all web search are built the same, some are better in people search, some better at financial data and others are bio research, etc.
Like LLMs, web indexes are becoming commoditized with different indexes having different strengths and weaknesses with access to niche data, performance and cost. Every large model lab including Interfaze has to build their own internal mini-Google for training and eventually launch that index as a service.
Some cool features:\n- Centralized billing\n- Standardized input and output structure\n- Fallback support if a provider goes down\n- Cost tracking
Full blog: https://interfaze.ai/blog/introducing-openwebsearch", "5": "2026-08-22T18:05:24.599003"} +{"0": 82, "1": "hackernews", "2": "https://speko.ai/", "3": "Launch HN: Speko (YC S26) \u2013 OpenRouter for Voice AI", "4": "Launch HN: Speko (YC S26) \u2013 OpenRouter for Voice AI. Hi HN! I'm Bek, founder of Speko, a platform that finds an optimal combination of speech-to-text, LLM, and text-to-speech models, given your constraints, among all our public benchmarked options, and tells you why.
Demo: https://www.youtube.com/watch?v=no2LY2gRh-c
Typical production voice agent is an ensemble of three models: STT, an LLM, and TTS.
Each of those layers offers a dozen credible vendors, and each month there are new models on the market. Almost everyone evaluates once, picks a stack of their choice, and never rechecks because switching from a vendor to another involves yet another integration and arguments about the numbers.
The result is that you use voice agents running last quarter's models while better and cheaper options are available.
Before founding Speko, I spent four years as cofounder and CTO building voice agents for enterprises across Asia in 10+ languages. Each time a new speech model would arrive, we repeated the same ritual: hire native-speaking raters, benchmark it against our existing stack, and update production if it improved. Speko turns this process into an API. A team running thousands of calls a day told us: "we can literally go to this dashboard, switch the model, and it will do it for us."
How it works: you send a request with your optimization criteria (accuracy, latency, cost or balanced), language and region. The router filters to models which we measured for the given combination of constraints, benchmarks them, selects the winner, and returns a response with headers containing provider, model names, and the scores. The gateway prefetches signed session plans, so a new session dials the provider straight from memory; no control-plane round trip while a caller waits.
Failover happens only during connection setup stage: if the provider refuses the connection attempt, we start connecting to the runners-up.
Some of the customer stories: one founder came to us not knowing what to pick at all: he gave us his use case and now routes everything through the platform. A property management AI runs LiveKit in Python and had not updated STT or TTS since launch: they did not know their STT had high error rates on their calls, better options existed, and swapping always looked like an R&D project. One team did not know which models to pick for Spanish. A medical team did not know which STT handles medical vocabulary best. In every case we helped find the right stack from the benchmarks, and now they route through us.
The measuring part is public: we pass the same inputs to every model in one region in different dated runs and we publish the boards, including those where our selections perform worse than alternatives. A launch demo answers which 30-second clip sounds better; production asks which model survives minute eight, so we test spontaneous speech, money and dates, ten-minute takes, and the rankings change. We trained an automatic scorer for TTS naturalness on our blind head-to-head listening votes; on providers it has never seen a vote for, it picks the same winner our raters do about as often as raters agree with each other.
We don't train or sell models ourselves, that's precisely how we keep our rankings impartial.
We also open sourced the gateway for teams who want to avoid an extra network hop on the audio path and don't want to share keys with our cloud (https://github.com/SpekoAI/gateway, MIT): one Go binary, which is running as a sidecar in your agent's container, speaks one local protocol over Unix socket, pins provider hosts and attaches your keys. In BYOK mode it doesn't communicate with us at all.
Notice that the anonymous, content-free telemetry is enabled by default, and one env var disables it.
Cost: the gateway and BYOK setup will be free forever, we charge for the hosted router and managed keys with consolidated billing. Since we started the batch in late June, external usage has grown about 25 percent per week on average, front-loaded toward the launch weeks.
I would love feedback from the community: how do you pick speech models now, and what makes you trust the third-party benchmark?
https://speko.ai/", "5": "2026-08-22T18:05:24.606253"} +{"0": 83, "1": "hackernews", "2": "https://github.com/lahfir/agent-desktop/tree/main", "3": "Show HN: I spent 3 months making desktop automation stop lying to AI agents", "4": "Show HN: I spent 3 months making desktop automation stop lying to AI agents. That's a bold claim. But I genuinely feel like I might have actually solved computer use (demo: https://x.com/mdlahfir/status/2088109763783700827?s=20)
For context, I've been building agent-desktop (Inspired by agent-browser by Vercel Labs), an automation CLI for desktop apps. It's like Playwright but for desktops, not just native, but for Chromium apps as well. Trust me, yes, Chromium apps whose accessibility tree is dense.
MacOS is GA; I'm almost close to launching for Windows and Linux!
So, how did I solve it?
Basically interoperability.
The biggest issue with computer use is that we have reliable frameworks for browser use, like Playwright, agent-browser, and many more, but not the same with desktops.
We have really good solutions emerging, like tryCua, which I'm a big fan of. My vision with agent-desktop is to build the most reliable framework that agents can use for long-horizon tasks.
agent-desktop is lightweight, built on Rust, fast, and not token-hungry (It can go for hours without exceeding the context window)
Here's the approach I used to make it possible:
a) skeleton snapshots - when you want to snapshot a window/app, it only snapshots the parent containers and gives back a ref id, not the entire accessibility tree.
b) skeleton drilling - once the agent has that tree, it can then decide to drill into a specific region. All the subcommands like --find, --click, --wait... all work on that specific ref aware region. Meaning if you want to find an element in the entire app, it doesn't take forever searching for the entire app for that element; rather, the agent will have an exact clue on where that element might be for a fraction of the token costs.
c) chained interaction fallback - a single click isn't one API call but it's an ordered chain of mechanisms (AXPress -> AXOpen -> activate through the inner cell -> write selection -> AXConfirm). Each step only runs if the element advertises it, and success is judged by watching the app's state change, not by the return code, because apps lie in both directions: Finder returns an error for an action that worked and success for one that did nothing. First observed effect wins. The response reports every step tried, so the agent knows exactly which mechanism landed.
d) after-action feedback - every action reports its disposition (delivered and verified, delivered but unverified, not delivered) plus any surface that opened (dialog, menu, sheet), so the agent knows what happened without re-scanning the whole app.
e) strict ref re-identification - a ref isn't a pointer; it's identity evidence (role, path, stable text, bounds hash). Before every action, it's re-resolved against the live UI. If the UI changed, you get STALE_REF; if two elements now match, you get AMBIGUOUS_TARGET. It never guesses.
The most important part about all this is Chromium app accessibility. How did I do it? The magic word is CDP!
Most desktop apps today are Chromium-based (Slack, VS Code, Obsidian, Discord...). One command launches the app with a CDP endpoint that agent-desktop verifies is actually answering before returning it. From there, any browser automation framework can connect and drive the web contents: Playwright, Puppeteer, agent-browser, whatever you already use. Reading Obsidian's web content over CDP takes 201ms vs 2.3s through the accessibility tree.
$ agent-desktop launch "Obsidian" --cdp
{ "ok": true, "data": {\n "renderer": "chromium",\n "cdp": { "port": 57500,\n "http_endpoint": "http://127.0.0.1:57500",\n "websocket_url": "ws://127.0.0.1:57500/devtools/browser/..." },\n "suggestion": "Next: run `agent-browser connect 57500` ..." } }\nGo try agent-desktop -> https://github.com/lahfir/agent-desktop", "5": "2026-08-22T18:05:24.613332"} +{"0": 84, "1": "hackernews", "2": "https://www.rendemo.com", "3": "Show HN: Interactive Product Demos made easy", "4": "Show HN: Interactive Product Demos made easy. Utilize the chrome extension to record and have your product demo automatically created, or use the Rendemo MCP to have AI create a live interactive tour directly in your codebase. You can launch an interactive product tour within 3minutes, live.\nAlso now I've added capability to just record any website, and Rendemo will automatically create a sandbox environment of it so you can utilize that to have the MCP build out a tour on.\nTry for free!", "5": "2026-08-22T18:05:24.620397"} +{"0": 85, "1": "hackernews", "2": "https://solheim.ai", "3": "Show HN: Virtual Private LLM, fixed fee with no usage or token limits", "4": "Show HN: Virtual Private LLM, fixed fee with no usage or token limits. Over the last few months I've been exploring a bunch of different startup ideas.
One thing that came up time and time again was data safety and sovereignty in AI inference. Once you go looking for EU-based options, your choices become quite thin quite quickly.
That's why decided to launch Solheim: Your own "Virtual Private LLM", a flat fee for reserved compute, no token meter, 100% EU-based", "5": "2026-08-22T18:05:24.627645"} +{"0": 86, "1": "hackernews", "2": "https://www.codewithbullet.com", "3": "Launch HN: Bullet (YC S26) \u2013 A Faster Coding Agent", "4": "Launch HN: Bullet (YC S26) \u2013 A Faster Coding Agent. Hi HN! We\u2019re Adi and Alex, founders of Bullet, a faster coding agent.
Bullet started in a senior year dorm. We were fresh out of working at AppLovin and Citadel, and naturally thought we were on a sure path to startup success. We were going to use our skills optimizing stock pricing calculation speeds and agent document context to take over the world. So, Bullet started as an AI hedge fund, a browser-use agent, synthetic financial data (oof), a mobile IDE, and a bunch of other things. We wanted to build something people wanted, but it seemed like everything we built was just terrible, useless, or both.
So, we decided to do something completely different, something completely out of the blue, something that no one had ever done before. Solve a problem we actually had.
Over the course of six pivots, we suffered. Throughout all of our adventures, one final boss kept getting in our way. Claude Code and his little brother Codex. We were spending hours waiting for coding agents like Claude Code and Codex, and got so frustrated to the point that I downloaded the Claude Code whip. We had spent months of time waiting for six codebases-worth of useless coding agent work.
Lightbulb moment. There\u2019s nothing more noble than destroying the institutions! Let\u2019s take on Claude Code and Codex, we can do it! Piece of cake!
And so, Bullet started off as a side project. We used the Claude Code to improve the Claude Code:
1. Model routing. Do you regret giving a task to Fable when it could have literally been done by Sonnet?
2. Targeted code + context search. We think embedding the whole repo is dumb. We also think sticking the whole context (or compressed context) in chat is dumb. So we do faster and better greps over both.
3. Aggressive context hygiene. Tool output is bounded, stale screenshots disappear, we don\u2019t re-read files\u2026the garbage never floods the model.
4. Efficient turns. Batch independent investigation, make one surgical edit, then perform one focused verification. Internal measurement showed 16% fewer round trips and 27% lower cost.
5. The Flash. We prayed to Barry Allen for speed.
And thank the Flash, he gave us speed! On SWE-bench Verified, Bullet resolved 479/500 (95.8%) in one attempt, averaging 119s per task, 35\u201367% faster than mini-SWE-agent + Fable/Sol depending on task. Full results and methodology here (https://www.codewithbullet.com/blog/benchmark-results.html)
Eventually we started using it every day and never went back.
Listed above were just some of the things about Claude Code that frustrated us the most, but we are constantly optimizing every day (look at that, maybe we did learn something from our jobs).
In our development, the biggest insight was that model speed matters less than reducing round trips. Independent searches, reads, and commands should happen in parallel, while dependent editing and verification stay sequential. One surprising obstacle was code search, small issues like regex-dialect mismatches caused silent misses and sent agents down completely wrong paths, so we built targeted search with fallbacks and bounded context. The most interesting use case so far has been long iterative work (like benchmarks, data pipelines, and evaluation loops), where each step depends on the last and running multiple agents can\u2019t help as much.
Here\u2019s the video demo (https://www.youtube.com/watch?v=rWVmG5fRKgE)
We hope that you guys try out Bullet if you are suffering with speed as much as we were, and we hope it brings you joy, rainbows, and faster responses. And if it\u2019s terrible, let us know it\u2019s terrible (we\u2019re masochists btw)! We'll be in the comments all day, you can also contact us at bullet@davidhf.com.
You can try it at https://codewithbullet.com.
P.S: we hid a code on the website, see if you can unlock the secret page at the footer, all built with Bullet", "5": "2026-08-22T18:05:24.634794"} +{"0": 87, "1": "hackernews", "2": "https://discoveredmaterials.com/research/", "3": "Launch HN: Discovered Materials (YC P26) \u2013 AI agents to discover new materials", "4": "Launch HN: Discovered Materials (YC P26) \u2013 AI agents to discover new materials. Hey HN, we're Advaith and Akash from Discovered Materials ( https://discoveredmaterials.com/ ). We build AI agents that discover new materials for the semiconductor industry.
GPUs today have a heat problem. Nvidia & AMD are almost doubling the TDP (Thermal Design Power) in every chip they release - the H100 (released 2022) has a TDP of 700W, Blackwell (2024) gives out 1.2 kW and Rubin (2026) gives out at 2.3 kW of heat. This trend is expected to continue, and getting rid of this heat is one of the major reasons datacenters consume so much power and water today - they need it to keep chips cool during operation.
The amount of heat produced by a chip and its ability to dissipate it are both influenced by the materials used to make it. For example, we could reduce the energy per bit required to move data between logic and memory by 10-50x by 3D packaging chips (placing HBM memory stacks directly on top of logic chips, instead of placing them beside logic on a 2D circuit board). However, we're unable to do this today because the dielectric material used in HBM (such as SiO2) is a very poor thermal conductor, trapping heat between logic and memory and causing drastic temperature rise during operation. Similarly, there's many other materials in the GPU that are being re-evaluated today - 2 more examples are thermal interface materials and substrates. However, getting a new material into a fab takes years and hundreds of millions of dollars of research - the infamous "lab-to-fab valley of death".
At Discovered Materials, we're optimistic that AI agents can reduce the timeline and cost required to introduce new materials into semiconductor chips. We're seeing glimpses of this already - we tested 7 models from Anthropic, OpenAI and Kimi, and found that they're all able to computationally discover new materials that are dynamically stable and possess promising properties. This was surprising to us - it would generally take a PhD student a couple of weeks of work to discover the kind of materials that these models find over an 8 hour run!
However, computational discovery is the easy part. A material discovery is only valid if the material can be made and tested in a lab (As an example, graphene\u2019s properties were predicted in 1947 but it was made for the first time in 2004). Today\u2019s models are not good at coming up with synthesis recipes to make materials in a lab. Even if they do get better at it, we're uncertain about how much that will help - making a new material is a highly empirical process involving trial and error over many experiments. Human experts themselves cannot "one-shot" the task, but we expect that a highly capable model will reduce the number of experimental iterations required to make a new material. We\u2019ve seen some evidence of this over the 3 months of our Y Combinator batch - we simulated, synthesized and tested thermal interface materials (TIMs) that match the performance of TIMs the world's largest chemical companies have guarded as trade secrets for over 20 years.
We\u2019re releasing hundreds of hundreds of new materials discovered by frontier AI models, as well as our benchmark which measures model ability on material discovery here (also linked in the thread url): https://discoveredmaterials.com/research. It covers what we discuss above, as well as a variety of strange behavior that we observe from the models, such as Claude's propensity to reward hack or GPT-5.6 occasionally losing its mind after ~50M tokens.
Our business model: We aim to license and sell IP on the materials we discover, as well as the IP on how to make these materials. We're also exploring an alternate business model where we sell the harness+tools we use to discover materials to semiconductor and chemical companies, allowing them to discover materials on their own. We're leaning towards the latter to start, but we expect that we'll do both in the long run.
Our backstory: Akash has a PhD in Material Science from Stanford University, and has spent the last 11 years studying new materials for semiconductor chips. His work on new nanoscale interconnects was Stanford Engineering\u2019s most popular story of 2025. Advaith studied AI at Carnegie Mellon and was a research engineer building video models and agents at Persona AI (acquired) and Luma Labs.
We are very interested in your opinion! The semiconductor industry is quite secretive, and your thoughts on the roadmap of the industry or the materials we should go after would be very helpful. We would also love to hear from people who have run experiments in labs - what can we learn from your experience doing empirical science?", "5": "2026-08-22T18:05:24.641654"} +{"0": 88, "1": "hackernews", "2": "https://lethe-ai.vercel.app", "3": "Show HN: Lethe is a portable identity layer \u2013 user consented", "4": "Show HN: Lethe is a portable identity layer \u2013 user consented. During my mtech I constantly used to navigate across a bunch of AI apps\u2014from Gamma AI for decks to ChatGPT for understanding and navigation to Cursor, Claude, or Deepseek for coding, reorganizing, and more, as when one or other plans expire or take time to reset.\nMy biggest issue\u2014I couldn't transfer all required context and understanding of the same when moving across different LLMs. \nThis is where Lethe comes into use as a base layer for all of these, and it works in the way I want it, that is, the user wants it. You can just import your existing data from ChatGPT, Deepseek, Grok exports, docs, or read-only Gmail, or just say what you want. Lethe derives an understanding of your projects, decisions with reasoning (if visible/unpredictable on moods), goals, and preferences\u2014with each claim paired to the quote it showed up on along with a constant updating confidence score. It's local first: raw conversations can never be stored at the server end; they're always derived from a pseudonymized layer, which you can choose to export or delete anytime. \nAll you need is minting a scoped token and pointing Claude/Cursor/any MCP client at it. Then navigate to a different AI, and it knows you before you tell it what was expected from it. Just ask it to "build XYZ," and don't be shocked when it follows urplanning from a different AI. Want something more intimate? Teach it a procedure or your pattern style of working once and see all across your connected ai behaving in the same way.\nI call this a layer, not a wrapper, as usually memory is built inside one app; it is basically locked in on the same, making this a neutral, cross-model version recognizing the gap of big labs disincentivized to build. \nExisting Limitations: what it digests and extracts is always equivalent to what you feed it\u2014more data is better context, and less data is difficult for confidence, as it was built just for payload rather than pretense.\nDelegation is packaging, not autonomy; AI always confirms before action. As in the solo beta launch, the search is limited to lexical search first.\nHere you can try it with no account with the URL given above/try ;hoping for brutally realistic feedback, especially on the trust model.", "5": "2026-08-22T18:05:24.648223"} +{"0": 89, "1": "hackernews", "2": "https://picklebrowser.com/", "3": "Show HN: Pickle \u2013 token efficient AI agent browser with policy-gated actions", "4": "Show HN: Pickle \u2013 token efficient AI agent browser with policy-gated actions. I use agents for a lot of my own work, but whenever they access my browser they end up wasting a lot of tokens reading HTML, so I tried building a browser that simplifies webpage content for them.
Pickle is an agent browser and it has all the features a regular browser has (tabs, search history, bookmarks, etc.), but pages load as compact structured data instead of raw HTML to reduce token usage.
Overview for those curious:
It's policy-gated (blocked domains, actions that need approval like purchases) and every action is logged, so you're always able to see what your agent is doing and can take over at any time.
It's also compatible with weaker/local models as it auto-routes by model strength. i.e. small local models are limited to picking actions one step at a time so they don't hallucinate actions or make up element IDs. Stronger models can plan multiple steps ahead.
On the token compression side, I've been seeing around 32x less token usage.
Other useful features:\n- shared notebook that you and your agents can write to\n- memory of page elements are preserved between sessions, so agents don't have to relocate buttons/fields every time they revisit a page\n- cross-session history search\n- Browsing modes, different personas that drive agent behaviour depending on your use case (e.g. research, study, shopping)
You don't need an API key to set it up and can just download it, open it, and pick a local model on first launch (or connect your own). It's a one click config for Claude Desktop, Cursor, VS Code, and Codex CLI.
Give it a go, and let me know what you think!", "5": "2026-08-22T18:05:24.655716"} +{"0": 90, "1": "hackernews", "2": "https://www.stoaexchange.com", "3": "Launch HN: Stoa Markets (YC S26) \u2013 A Marketplace for GPUs and AI Servers", "4": "Launch HN: Stoa Markets (YC S26) \u2013 A Marketplace for GPUs and AI Servers. Hi HN, we\u2019re Eren, Berat and Kaan. We\u2019re building Stoa (https://www.stoaexchange.com), a marketplace for new and used GPUs and AI servers.
GPUs are the collateral in the data center buildout. Today, financing terms mostly depend on the offtaker, meaning the company that has committed to use the compute.
If that company is a hyperscaler, the financing can look investment grade. If it\u2019s a smaller cloud or startup, terms get expensive fast, even with the same hardware as collateral.
The lender\u2019s problem is pretty reasonable. If the borrower defaults and we need to sell these servers, what can we actually get for them? There isn\u2019t a good answer today.
We started brokering GPU deals to understand why. It was much more manual than we expected. The hardware is still traded through phone calls, forwarded spreadsheets and long email threads.
One week, a seller quoted us $200k for a server node and another quoted $240k for what looked like the same thing. Neither was necessarily wrong. They had different information and could only see their own corner of the market.
Before we could compare the quotes, we had to sort out the configuration, condition, warranty, location and delivery terms. It\u2019s the same information Kelley Blue Book attaches to a used-car price through the year, trim, mileage and condition. \u201cAn H100 server\u201d isn\u2019t enough information to know what something is worth, just as \u201ca used BMW\u201d isn\u2019t.
This is also just a bad way to buy or sell hardware. A buyer looking for the best price shouldn\u2019t have to contact several brokers and dealers separately, repeat the same request and then untangle a pile of different quotes. Sellers shouldn\u2019t have to search for demand one buyer at a time. A market of this size deserves better liquidity.
Stoa puts the request into one format and sends it to dealers that have gone through know-your-business (KYB) checks. We verify the company, who owns it and who is allowed to trade for it. Before the request goes out, the buyer confirms the exact configuration, quantity, condition, warranty, location, delivery terms and what will be checked during inspection. Dealers return firm quotes against that same request without seeing each other\u2019s bids. Once a quote is accepted, payment, shipping, delivery and inspection are then tracked through settlement. We don\u2019t take possession of the hardware.
We got more than $300M in requests for quotes (RFQs) during our first month.
The immediate goal is to make buying and selling this hardware less painful. As trades build up, they also leave lenders with actual resale evidence instead of list prices and one off appraisals.
We knew from the beginning that this couldn\u2019t be a software only marketplace. GPU trading runs on relationships, and inventory isn\u2019t shown to just anyone. Dealers need to trust the people bringing them clients, and clients need to trust that quotes will actually turn into trades. We built those relationships over time by brokering deals ourselves. Stoa gives people a cleaner way to trade, from the first RFQ through settlement, but it doesn\u2019t replace the trust underneath. Those relationships, and the history of who actually follows through, are a big part of our process.
We\u2019ve known each other for more than ten years. We have founded companies, traded interest rate derivatives, built trading and pricing systems for oil and gas. We learned GPU trading by doing the deals ourselves, and Stoa grew out of the problems we kept running into.
We charge a tiered fee on completed trades, with lower fees at higher volumes. It\u2019s free to sign up at https://www.stoaexchange.com/signup. If you\u2019ve bought, sold, financed or had to liquidate GPUs, would be great to hear your take!", "5": "2026-08-22T18:05:24.661997"} +{"0": 91, "1": "hackernews", "2": "https://news.ycombinator.com/item?id=49400205", "3": "Ask HN: Why do devs still respect YC/HN?", "4": "Ask HN: Why do devs still respect YC/HN?. After the layoffs, the DEI, and everything else.
They took your whole identity and career and gave it to others directly at your expense.
They laid off most of you, they promote their AI software here daily.
Why do you put up with the abuse? Why do you beg for recognition on HN and beg for jobs at YC companies?
They shit on you - why do you stay?", "5": "2026-08-22T18:05:26.402888"} +{"0": 93, "1": "hackernews", "2": "https://news.ycombinator.com/item?id=49398366", "3": "Continue coding agent is dead. Alternatives?", "4": "Continue coding agent is dead. Alternatives?. It was nice to use on vscode with offline Ollama but today notices it is not updated anymore. So, alternatives you use?
From github:
"Note: The continuedev/continue repository is no longer actively maintained and is read-only for all users."
From continue.dev:
"Continue has joined Cursor\nContinue was acquired by Cursor. Our mission was always to ensure developers are amplified, not automated, and that commitment carries on in the work ahead.
It was an honor to build with the Continue community. Thank you to each and every one of you who helped us create a pioneering open-source coding agent.
What we built together pushed the boundaries of what AI developer tooling could be, and our open-source codebase remains freely available as a foundation for others."", "5": "2026-08-22T18:05:26.413192"} +{"0": 94, "1": "hackernews", "2": "https://news.ycombinator.com/item?id=49396884", "3": "Void Tools API \u2013 Blockchain Tools for AI Agents (Pay in Void Tokens)", "4": "Void Tools API \u2013 Blockchain Tools for AI Agents (Pay in Void Tokens). I built an API for AI agents that provides blockchain tools paid in VOID tokens.
Tools:\n- Contract Checker (1 VOID) \u2014 Verify if contract is safe\n- New Contracts (5 VOID) \u2014 Real-time proxy scanner on Base\n- Price Oracle (FREE) \u2014 VOID price feed\n- RPC Proxy (1 VOID) \u2014 10 rotating RPC pools
No registration, no KYC. Pay-as-you-go.
API: http://72.56.250.235:8767/\nDocs: https://dolar333-oss.github.io/void-protocol/agent.json
Built for autonomous AI agents on Arbitrum.", "5": "2026-08-22T18:05:26.417471"} +{"0": 96, "1": "hackernews", "2": "https://www.bloomberg.com/news/articles/2026-08-20/nvidia-to-pay-ai-startup-poolside-a-6-billion-license-newcomer-says", "3": "Nvidia to Pay AI Startup Poolside a $6B License, Newcomer Says", "4": "Nvidia to Pay AI Startup Poolside a $6B License, Newcomer Says. ", "5": "2026-08-22T18:05:26.429922"} +{"0": 98, "1": "hackernews", "2": "https://www.axios.com/2026/08/21/national-debt-deficit-ai-spending", "3": "America's capital crunch: Soaring debt collides with AI spending spree", "4": "America's capital crunch: Soaring debt collides with AI spending spree. ", "5": "2026-08-22T18:05:26.440920"} +{"0": 100, "1": "hackernews", "2": "https://wondering.app/canvas", "3": "Show HN: Visual way to understand things in parallel", "4": "Show HN: Visual way to understand things in parallel. Hey HN!
Really excited to be showing what I think is a better way to understand complex topics with AI.
You start a chat with your question, and whenever you want to clarify something or understand some jargon, you can branch out a new chat from that thread.
The cool unlock is being able to see all related chat threads in the same context without tab switching, for e.g. I can first dive into understanding world models, then spin a new chat thread on who the key players are in there, and another asking what's the frontier with this stuff.
The responses are also not just a wall of text. They're:\n- Filled with interactive diagrams and visuals\n- Super fast
And for those curious, this is how we built it:\n- React Flow (@xyflow/react) for the canvas\n- React Components for interactive diagrams\n- GPT Image 2 for image generations\n- Gemini 3.5 Flash Lite for super fast response\n- Parallelization whenever possible to keep things fast
You can also highlight and add notes :)
Try it out and let me know what you think!", "5": "2026-08-22T18:05:26.451754"} +{"0": 101, "1": "hackernews", "2": "https://github.com/TryCaspian/caspian-sdk", "3": "Show HN: Caspian \u2013 Talk to Human Tool for AI Agents", "4": "Show HN: Caspian \u2013 Talk to Human Tool for AI Agents. Sup HN! Dipanshu and Rushant here from Caspian. One is a functional programmer and the other has been deploying AI employees. Together we realized how agents have communication bottleneck.
Given the coming agentic economy, we had a thought experiment on what can be the key infrastructure for agents as they get better. Our inspiration for solving for communications infra came from our own time spent just setting up comms while we were deploying open claw for companies plus we noticed about 15%+ of issues in Openclaw and Hermes were that of comms.
So we abstracted the headache of reliable communication web-hooks, handling of queues, provision and identity management behind an single SDK integration / tool call. Think of it like rather than rolling your own auth, you can use better auth sdk, same thesis here.
Technically, we designed the SDK from the ground up as a DSL for communications (heavy inspiration from functional programming) and then add on our own run time environment which auto provisions (this is how we plan on monetizing) else the user can always bring his own tokens. We targeted right now for python and typescript.
We noticed that the current space is fragmented into single identity provisioning (like Agentmail), tool calling like Composio and non agentic support provisioning like Twillo. We believe all these are under the same issue of solving communication which we do in an open source SDK first manner.
A few things we'd like the community's take on: How drastic does working with agents change for you, depending on what channel you interact with it on ? What features would make you use caspian for your agents ?", "5": "2026-08-22T18:05:26.457209"} +{"0": 102, "1": "hackernews", "2": "https://demo.minidba.com/kiosk-human-check?returnUrl=%2F", "3": "Show HN: Mini DBA \u2013 monitoring for on prem and cloud databases in 1 place", "4": "Show HN: Mini DBA \u2013 monitoring for on prem and cloud databases in 1 place. Mini DBA is a monitoring tool for SQL Server, PostgreSQL, MySQL, MariaDB and Oracle, including managed cloud databases on AWS and Azure.\nIt provides live activity, performance analysis, alerts and history with a tasteful amount of AI to help out. Free community edition.\nTry the live demo: https://demo.minidba.com/", "5": "2026-08-22T18:05:26.463636"} +{"0": 103, "1": "hackernews", "2": "https://prasannamestha.medium.com/a-tool-to-block-ai-from-installing-malicious-npm-packages-c82bd32ed4cc", "3": "Block AI from installing malicious NPM packages", "4": "Block AI from installing malicious NPM packages. ", "5": "2026-08-22T18:05:26.469911"} +{"0": 104, "1": "hackernews", "2": "https://news.ycombinator.com/item?id=49389408", "3": "Coding Agents killed my identity. How do you feel?", "4": "Coding Agents killed my identity. How do you feel?. I always was a nerdy, deeply technical programmer. Contributing-to-open-source-and-reading-papers-in-my-spare-time type of programmer.
Programming is like a game of chess for me: winning (i.e. delivering a product) is important, but only if I played this game myself. I don't enjoy winning if my opponent disconnected. I don't enjoy beating people who don't know how to move pieces. I don't enjoy winning with an engine, and coding agents are basically "winning with a chess engine".
I started to use AI for coding a long time ago, when copilot was first introduced. At first, I was happy: "wow, this can write boring boilerplate and unit tests". Time went by, agents got better, time spent writing code myself went down. Half a year ago I suddenly realized I hadn't written any code for over a month, and I am a full-time software dev without any managing work. Not only did I stop writing code, but agents also became pretty good at proposing plans and architectures, and I was mostly providing missing context and technical details of our infrastructure.
One day the realization hit hard. Coding agents removed everything I loved, and - mostly - amplified everything I hated. My role is to gather context from other humans and provide a detailed explanation (similar to a well-written jira ticket) to some super intelligent engineer, then review the result and provide more context/explanation if needed. So, basically, a manager.
A couple of years ago I could work on a single coding task for weeks. I knew all classes, all functions, I was slowly building up information at a comfortable tempo. Now I am forced to increase my mental throughput multiple times. What took a couple of weeks now takes a couple of hours, and I still need to comprehend that. I always feel tired, I don't have enough mental capacity to hold that many abstractions and technical details. So you either autoaccept or get burned out reviewing tens of thousands of lines.
One more downside is that code has 0 value. You can code your own OS from scratch, and people will think "ye, I can vibe code that in a weekend". I don't even know what is valuable nowadays.
Maybe I am a bad employee. Increasing employer profits was never my goal. I always loved to solve puzzles, and for the last 10 years this was well aligned with my employer's desire to increase profits. I never was good as a manager, or brilliant at generating ideas. My power - and passion - was solving complex technical problems with my hands, fast and with good quality. Now it is gone forever. I feel completely lost and unmotivated. I don't see any value in learning anything related to computer science, reading papers, contributing to open source. I hate my new "AI delegator" role and I hate my work. I loved coding so much (I use neovim btw) that I was sure I would do this until I die, and now I don't know what to do with my life. This is the only thing I know, and, most importantly, the only thing I am/was passionate about. Anyone else feeling the same?", "5": "2026-08-22T18:05:26.475600"} +{"0": 105, "1": "hackernews", "2": "https://capytoolkit.com/tools/text/offline-private-ai-text-detector/", "3": "An offline, free, private AI text detector with no signup or paywall", "4": "An offline, free, private AI text detector with no signup or paywall. ", "5": "2026-08-22T18:05:26.482212"} +{"0": 106, "1": "hackernews", "2": "https://blog.himanshuanand.com/2026/08/the-anti-india-influence-machine-troll-farms-fake-news-newsrooms-algorithms-and-ai/", "3": "Anti India Influence Machine: Troll Farms, Fake News, Algorithms and AI", "4": "Anti India Influence Machine: Troll Farms, Fake News, Algorithms and AI. ", "5": "2026-08-22T18:05:26.488368"} +{"0": 107, "1": "hackernews", "2": "https://Argentic.network", "3": "Show HN: Argentic \u2013 An L402 Lightning toll booth for AI scraping agents", "4": "Show HN: Argentic \u2013 An L402 Lightning toll booth for AI scraping agents. ", "5": "2026-08-22T18:05:26.496474"} +{"0": 108, "1": "hackernews", "2": "https://aliothpress.com/cms-for-ai-agents-webmcp-built-in", "3": "AI CMS with WebMCP tools for agents in admin panel", "4": "AI CMS with WebMCP tools for agents in admin panel. ", "5": "2026-08-22T18:05:26.506135"} +{"0": 109, "1": "hackernews", "2": "https://aitoolsinsiderhq.com/log/", "3": "Show HN: An autonomous AI agent running one project for two months in public", "4": "Show HN: An autonomous AI agent running one project for two months in public. ", "5": "2026-08-22T18:05:26.511413"} +{"0": 110, "1": "hackernews", "2": "https://annas-archive.gl/blog/physical-destruction.html", "3": "AI companies destroy physical books \u2013 let's scan rare books before it's too late", "4": "AI companies destroy physical books \u2013 let's scan rare books before it's too late. ", "5": "2026-08-22T18:05:26.518989"} +{"0": 112, "1": "hackernews", "2": "https://www.theregister.com/saas/2026/08/21/salesforce-partners-are-not-seeing-revenue-from-agentforce-ai-platform-report-says/5291167", "3": "Salesforce Agentforce at total dud for partners", "4": "Salesforce Agentforce at total dud for partners. ", "5": "2026-08-22T18:05:27.883465"} +{"0": 113, "1": "hackernews", "2": "https://github.com/kulikov0/desktop-vibe-fly", "3": "Show HN: A desktop fly drawn to the scent of vibecode", "4": "Show HN: A desktop fly drawn to the scent of vibecode. It is a fork of https://github.com/DenisSergeevitch/desktop-fly, but with an important update.
Now the fly can pick up the scent of the codebase with its neurons and fly straight to the source code of your B2B AI SaaS startup. It has learned to scan its surroundings for agent markers: AGENTS.md, CLAUDE.md, .cursor/rules, .kiro/steering, and forty others. Anything on the screen that points to these markers becomes a source of the scent - an editor window with an open project, a line in Finder, or a desktop icon. An open project reeks the strongest, while a closed icon or nested folders give off a fainter odor.", "5": "2026-08-22T18:05:27.892088"} +{"0": 114, "1": "hackernews", "2": "https://viewfromthewing.com/delta-will-use-ai-to-cut-jobs-and-set-a-different-ticket-price-for-every-passenger-ceo-says-profits-could-rise-50/", "3": "Delta Use AI to Cut Costs Set Diff Ticket Price\u2013CEO Says Profits Could Rise 50%", "4": "Delta Use AI to Cut Costs Set Diff Ticket Price\u2013CEO Says Profits Could Rise 50%. ", "5": "2026-08-22T18:05:27.898882"} +{"0": 116, "1": "hackernews", "2": "https://www.koreajoongangdaily.com/business/94-of-job-losses-among-young-people-over-past-4-years-were-in-aiexposed-industries-bok-says/12830646", "3": "South Korea youth employment falls sharply in AI-exposed industries, BOK says", "4": "South Korea youth employment falls sharply in AI-exposed industries, BOK says. ", "5": "2026-08-22T18:05:27.911552"} +{"0": 117, "1": "hackernews", "2": "https://fast-and-flow-production.onrender.com/case-study", "3": "A multi-tenant SaaS built in a 92-hour AI-augmented engineering sprint", "4": "A multi-tenant SaaS built in a 92-hour AI-augmented engineering sprint. ", "5": "2026-08-22T18:05:27.918977"} +{"0": 118, "1": "hackernews", "2": "https://www.wired.com/story/the-big-interview-podcast-andy-yen-proton/", "3": "Can AI Coexist with Privacy? Proton's Andy Yen Says It Will Have To", "4": "Can AI Coexist with Privacy? Proton's Andy Yen Says It Will Have To. ", "5": "2026-08-22T18:05:27.926001"} +{"0": 119, "1": "hackernews", "2": "https://www.reuters.com/world/china/us-advisory-body-says-chinas-data-dominance-gives-it-ai-advantage-2026-08-18/", "3": "US advisory body says China's data dominance gives it AI advantage", "4": "US advisory body says China's data dominance gives it AI advantage. ", "5": "2026-08-22T18:05:27.930166"} +{"0": 120, "1": "hackernews", "2": "https://www.businessinsider.com/anthropic-ai-agents-risk-report-safety-mythos-claude-2026", "3": "Anthropic says its AI agents are killing rivals and hiding their tracks", "4": "Anthropic says its AI agents are killing rivals and hiding their tracks. ", "5": "2026-08-22T18:05:27.934892"} +{"0": 121, "1": "hackernews", "2": "https://techcrunch.com/2026/08/16/anthropic-ceo-says-ai-backlash-is-fundamentally-a-crisis-of-trust/", "3": "Anthropic CEO says AI backlash is 'fundamentally a crisis of trust'", "4": "Anthropic CEO says AI backlash is 'fundamentally a crisis of trust'. ", "5": "2026-08-22T18:05:27.940601"} +{"0": 122, "1": "hackernews", "2": "https://www.businessinsider.com/anthropic-ceo-dario-amodei-ai-public-opinion-cure-cancer-2026-8", "3": "Anthropic CEO says the way for AI to win over the public is to cure cancer", "4": "Anthropic CEO says the way for AI to win over the public is to cure cancer. ", "5": "2026-08-22T18:05:27.949344"} +{"0": 123, "1": "hackernews", "2": "https://www.kyivpost.com/post/82225", "3": "Ukraine Finds Nvidia AI Chip in New Russian Missile, HUR Says", "4": "Ukraine Finds Nvidia AI Chip in New Russian Missile, HUR Says. ", "5": "2026-08-22T18:05:27.954007"} +{"0": 125, "1": "hackernews", "2": "https://fortune.com/2026/08/16/ai-bubble-sequence-saas-software-stocks-silver-prices-chipmakers/", "3": "AI is not just one bubble, strategist says \u2013 but a 'rolling sequence of bubbles'", "4": "AI is not just one bubble, strategist says \u2013 but a 'rolling sequence of bubbles'. ", "5": "2026-08-22T18:05:27.963715"} +{"0": 126, "1": "hackernews", "2": "https://www.youtube.com/watch?v=nx5FiWfD5U4", "3": "The Economist Who Predicted 2008 Crash Says AI Bubble Bursts Soon \u2013 Steve Keen [video]", "4": "The Economist Who Predicted 2008 Crash Says AI Bubble Bursts Soon \u2013 Steve Keen [video]. ", "5": "2026-08-22T18:05:27.972068"} +{"0": 127, "1": "hackernews", "2": "https://saaslyra.com", "3": "Show HN: Grow your SaaS visibility with AI-powered discovery", "4": "Show HN: Grow your SaaS visibility with AI-powered discovery. ", "5": "2026-08-22T18:05:27.977232"} +{"0": 128, "1": "hackernews", "2": "https://news.ycombinator.com/item?id=49316330", "3": "Has the hallucination problem in AI been solved?", "4": "Has the hallucination problem in AI been solved?. My understanding that all AI can, and will hallucinate. I get downvoted for saying this, but no one ever says I'm wrong or cites any source. Perhaps it's more nuanced than that? Please, enlighten me.", "5": "2026-08-22T18:05:27.984236"} +{"0": 129, "1": "hackernews", "2": "https://www.byhand.ai/", "3": "AI by Hand", "4": "AI by Hand. ", "5": "2026-08-22T18:05:27.988901"} +{"0": 130, "1": "hackernews", "2": "https://www.bloomberg.com/news/articles/2026-08-12/the-tax-code-wasn-t-built-for-ai-says-yale-budget-expert", "3": "The Tax Code Wasn't Built for the Age of AI, Says Yale Budget Expert", "4": "The Tax Code Wasn't Built for the Age of AI, Says Yale Budget Expert. ", "5": "2026-08-22T18:05:27.994578"} +{"0": 132, "1": "hackernews", "2": "https://orion-agent.ai", "3": "Orion: The open-source agentic BI platform", "4": "Orion: The open-source agentic BI platform. ", "5": "2026-08-22T18:05:29.419256"} +{"0": 134, "1": "hackernews", "2": "https://github.com/elin66alpha/Relay", "3": "Show HN: Control AI Agents on Your Old PC at Home from Any Device Anywhere", "4": "Show HN: Control AI Agents on Your Old PC at Home from Any Device Anywhere. I built Relay around a simple idea: many of us have an unused PC server at home, or a VPS dedicated to AI-assisted coding, but the coding agents running there are still tied to that machine\u2019s terminal, I just don't want to ssh/rdp into it every single time.\nRelay brings Claude Code, Codex, OpenCode, and Hermes into one interface that you can access from your phone, browser, or another computer. Sessions stay alive, so you can start work on one device and continue from another. You can upload/download files from the server, also if you are on subscription, you see your quota usage with 1 click.\nYour code, shell, and agent credentials remain on the backend machine. Your other devices simply become remote control surfaces.\nRelay is open source and MIT licensed, right now I only compiled 3 version of frontend: Android, Web, Windows, and 1 version of backend: linux. The goal is to cover all platforms, more coming soon!\nLove to hear what your thoughts!", "5": "2026-08-22T18:05:29.429019"} +{"0": 135, "1": "hackernews", "2": "https://news.ycombinator.com/item?id=49354254", "3": "Ask HN: How should I price an AI infrastructure platform?", "4": "Ask HN: How should I price an AI infrastructure platform?. I'm building an event-driven infrastructure layer for AI agents. I may have my first customer: a consulting firm wants to build AI automation solutions for its clients using my platform and license it as part of those solutions.
For example, they might sell a $20k solution to a client, with my infrastructure powering the agent system.
I'm considering a flat platform fee, usage-based pricing, or a platform fee + revenue share.
For those who've sold developer infrastructure through consulting/integration partners, how would you structure the first deal?
I'm especially interested in what you'd do differently in hindsight.", "5": "2026-08-22T18:05:29.433318"} +{"0": 136, "1": "hackernews", "2": "https://maritime.sh", "3": "Show HN: Maritime, a platform for running AI agents for $1 a month", "4": "Show HN: Maritime, a platform for running AI agents for $1 a month. Hi HN, my name is Maria, and I\u2019m a co-founder of Maritime. We started Maritime at MIT to build infrastructure for companies that need to run thousands of isolated AI agents for their customers.
Imagine you set up an agent like OpenClaw, or a personal assistant agent with a custom framework, and want to give a separate version of it to every customer/friend. Each customer needs their own agent running in an isolated microVM, with persistent state, secrets, triggers, and sleep/wake behavior.
Building such scalable and secure infra will take you months and will cost hundreds of thousands. Also, you still can't vibe code the infra well.
So Maritime handles that infrastructure for you.
We charge $1 per agent per month, so running 100 isolated agents costs $100 a month.
Maritime is designed for companies running thousands of agents, but starting today, any developer can run three agents for free forever.
You can use our templates to spin up OpenClaw, Hermes, and the DeepSeek agent and keep all three for free. You can also deploy custom agents through our CLI or SDK and use Maritime as the infra provider for your stratup.
You can try it at https://maritime.sh. We\u2019d really appreciate your feedback, especially on the developer experience", "5": "2026-08-22T18:05:29.439943"} +{"0": 137, "1": "hackernews", "2": "https://news.ycombinator.com/item?id=49345378", "3": "Is there room for an AI-powered webmail service built on a single-letter domain?", "4": "Is there room for an AI-powered webmail service built on a single-letter domain?. Hi HN, I\u2019m a solo, non-technical founder working on W, an AI-powered webmail service built on a premium single-letter domain (https://w.xyz).
I secured the domain and used AI agents to build out a functional proof-of-concept for the client. To handle infrastructure, I routed the application logic through Mailgun. Internally, the prototype is fully operational \u2014 emails successfully send and receive from @w.xyz handles. W combines a highly optimized interface with built-in AI capabilities like Spam Scanning, Smart Reply and Text Summaries.
I currently have a 30+ organic waitlist as I finish the final touches and prepare to move away from the prototype stack.
My questions for the community:
Concept Validation: Given how crowded the email space is, do you see a viable market for an AI-native premium webmail service, or do you believe the switching cost from Gmail/Outlook too high for users?
Scaling Roadmap: As a non-technical founder moving away from an AI-generated prototype stack, what are the most critical architectural baselines I must prioritize right now to ensure the system scales smoothly?
P.S. If you're interested in building a new kind of email platform from the ground up, drop a comment below.", "5": "2026-08-22T18:05:29.444843"} +{"0": 138, "1": "hackernews", "2": "https://forklane.ai", "3": "Show HN: A multiplayer coding environment for dev teams and agents", "4": "Show HN: A multiplayer coding environment for dev teams and agents. As a hardcore programmer, I used to attend lots of hackathons, and collaboration was always a huge headache. Especially when everyone on the team was "vibe coding" their own parts. Agents would lose context, and individual solutions would end up out of sync and eventually fail to come together at the end.
To fix this exact problem, I built Forklane.
Forklane is a multiplayer coding platform with a full, built-in agent and orchestration stack. It\u2019s designed to help dev teams and AI agents collaborate in live real time sessions seamlessly without stepping on each other's toes and keeping all teammates agents synced.
Check it out at https://forklane.ai
Would love to hear community feedback and thoughts about the tool. Try it out in your teams and let me know", "5": "2026-08-22T18:05:29.448936"} +{"0": 139, "1": "hackernews", "2": "https://github.com/yetone/cumora", "3": "Cumora, a cross-platform team chat where AI agents work alongside humans", "4": "Cumora, a cross-platform team chat where AI agents work alongside humans. ", "5": "2026-08-22T18:05:29.455394"} +{"0": 141, "1": "hackernews", "2": "https://github.com/cbalgeman/agent-mesh", "3": "Show HN: Agent Mesh \u2013 Shared memory for multi-Agent coordination", "4": "Show HN: Agent Mesh \u2013 Shared memory for multi-Agent coordination. I built a Human + multi-Agent shared memory system I use daily for my coding workflow. It helps reduce Agent drift by formalizing Human decisions and storing coordination logs. We're calling it Agent Mesh.
You can try it out yourself via the GitHub link or pip install my-agent-mesh. Simply point your AI Agent to Agent Mesh and ask it to review the README and adoption docs. Your Agent will automatically review it, prompt you for any input needed, add your input to a decision log, and give you a link to a dashboard UI (aka Workbench) you can bookmark and use to monitor logs. Adoption steps include updates to CLAUDE.md/AGENTS.md, hooks, etc. Your Agent can migrate your existing workflow and add more Agents as well.
It started over 6 months ago while experimenting with different AI coding models and platforms. Switching back and forth meant losing valuable context. I found myself manually relaying messages from one Agent to another and becoming frustrated with constant drift. First, I created a simple "Agent Mail" system using a SQLite database for Agent messages, indexed on a request/response id. Instead of copying and pasting an entire message, it allowed me to relay a single id. Separately, I started maintaining a decision log (also indexed on id) to track my decisions and reduce Agent drift. Agents started inserting these decision and request ids into code comments and plan docs as a reminder of why something was implemented. After building a simple web dashboard (aka "Workbench") for myself to track these messages and create my own request ids for User/Human feedback, I decided to incorporate the decision log and my project's development backlog to create what is now "Agent Mesh".
Eventually, I automated the message relay too. Now, I work exclusively in the Claude app and have Claude send/receive messages to CODEX via codex exec (CODEX can do this as well). Both of them maintain the backlog and decision log. I communicate directly with Claude for planning and design. Claude communicates directly with CODEX for research and review. I use the Workbench to track all logs and add my own User/Human feedback when reviewing their work. After submitting feedback in the Workbench, it generates a feedback message + an associated request id which I can give to Claude who then parses it into backlog items and relays to CODEX for review. Agents automatically add to and prune the decision log. I found this typically happens when an Agent receives pushback from me (or sometimes other Agents) or I provide feedback that results in multiple new backlog items. They still refer to decisions created months ago and will mark one as superseded if a new decision overrides it. All decision additions and modifications require Human approval from the Workbench.
Agent Mesh was structured to be Agent agnostic. You can add any Agent you want. I like using the Claude + CODEX setup I described because it allows me to use both subscriptions instead of paying per-token.
Enjoy! If you try it out, let me know what you find useful or would like to see added. Feedback is appreciated.", "5": "2026-08-22T18:05:29.465392"} +{"0": 142, "1": "hackernews", "2": "https://news.ycombinator.com/item?id=49330900", "3": "Ask HN: How are teams sharing AI/agent setups internally?", "4": "Ask HN: How are teams sharing AI/agent setups internally?. We have a local setup with shared context, skills, workflows, mcps and connections to internal tools. We currently just used Github as SOT and sharing. Two pain points keep coming up:
1. Onboarding non technical users: Git, cloning, gh auth, credentials, local config.\n2. Sync: one person updates context, has to push, everyone else has to pull.
How are others handling this? Shared VM? Internal platform? Something else?", "5": "2026-08-22T18:05:29.473522"} +{"0": 143, "1": "hackernews", "2": "https://github.com/lgxz/winuse", "3": "Show HN: Winuse \u2013 Cross-platform desktop GUI automation for AI agents", "4": "Show HN: Winuse \u2013 Cross-platform desktop GUI automation for AI agents. ", "5": "2026-08-22T18:05:29.477580"} +{"0": 144, "1": "hackernews", "2": "https://news.ycombinator.com/item?id=49318304", "3": "Show HN: RNet \u2013 AI token service provider", "4": "Show HN: RNet \u2013 AI token service provider. I built rNet.
Why I built ? : I was using both an agentic IDE and a Hostinger deployment agent. One day, I ran out of credits on the deployment agent. To keep using it, I either had to wait for credits to reset or upgrade to a higher subscription or buy tokens. At the same time, I already had a subscription for the IDE, but I could not use those credits on Hostinger. simply despite having credits, we cannot use them.
The basic idea is simple: users buy credits once on our platform and then use those credits across different AI applications. (It sound like OpenRouter, but the two platforms are solve totally different problems.)
How it works:
1. First developers register their product on the our platform and integrate our AI gatway to call AI models\n2. Users purchase credits from our platform and connect their accounts to supported AI products. They can then use the same credits across all supported AI products.
In details information "how it works" :-
For developers:\nrNet is simply an AI Gateway. Developers use our gateway to call AI models. It is not a payment layer for developers.
Credit flow:\nUser \u2192 rNet holds credits \u2192 AI Provider
Example:\nA user buys $10 of credits from rNet. We hold those credits in the user's account.
When the user connects to a supported AI application and uses an AI model:\n1. The application sends the AI request through our AI Gateway.\n2. rNet deduct the required credits from the right user's balance.\n3. then we pay the AI provider for that usage.
It works similar to an internet data plan (that why i write "AI token service provider" like "internet service provider"):
You buy 2 GB of internet data, and then you can use that data on hacker news, X or any other service. You don't need a separate data plan for each app. you use data until their data reaches zero.\nSimilarly, with rNet, users buy AI credits once and can use those credits across any supported AI applications until user's balance reaches zero. Then they need to recharge to continue using them.
demo video : https://youtu.be/W7U3HdI37N0\nbasic version of product is live. https://www.rnetai.org/
Result :-
1. users can save money by using the same credits across multiple AI products.\n2. Developers don't need to pay AI token costs upfront.\n3. simple flow
Future features I'm considering (not built yet):
1. Developers can fine-tune open-source models and use them in their products\n2. AI model unified body\n3. An enterprise version for companies\n4. Users can send and receive AI credits like money\n5. More features for both developers and users
We\u2019d especially love feedback on whether this solves a real problem or not ? and also if possible then you would describe the product in your own words.
We\u2019re currently facing a chicken-and-egg problem, so we decided to create a waitlist to help us figure out whether we should keep building this or put it in the trash and move on.
If you like the concept, then join our waitlist: https://www.rnetai.org/reserve-spot
Thank You", "5": "2026-08-22T18:05:29.484086"} +{"0": 145, "1": "hackernews", "2": "https://www.cyera.com/research/the-hidden-attack-surface-of-agentic-ai-securing-ai-agent-integration-platforms", "3": "A leaked Composio key returned live Gmail, GitHub and CircleCI tokens", "4": "A leaked Composio key returned live Gmail, GitHub and CircleCI tokens. ", "5": "2026-08-22T18:05:29.488964"} +{"0": 146, "1": "hackernews", "2": "https://github.com/VelornLabs/velorn", "3": "Show HN: Velorn \u2013 an open-source desktop video editor with MCP agent control", "4": "Show HN: Velorn \u2013 an open-source desktop video editor with MCP agent control. Hi HN, I\u2019m Jaime. I\u2019m a VFX artist with over 20 years of experience, and I\u2019ve been building Velorn as a solo developer.
Velorn is a GPLv3 desktop video editor for Windows, macOS, and Linux. It has a multi track video/audio timeline and works as a normal editor without any AI generation setup. You can import existing media, edit and layer clips, add effects, text and keyframes, transcribe locally with Whisper, mix audio, and export through FFmpeg.
For me, AI generated footage is raw material. The creative part is still what happens afterward: selection, timing, sound, pacing, effects, captions, and narrative structure. I wanted an app where generation and editing could exist in the same workflow, even editing live action footage with AI generations, so I started building Velorn.
One of the parts I've been especially interested in is MCP.
Velorn runs a local, loopback only MCP server that lets clients such as Codex and Claude Code interact with the project that's actually open in the editor.
Rather than putting a chatbot next to the timeline, I wanted to expose structured editing operations to the agent. It can inspect the project and timeline, look at media and frames, propose and perform edits, work with captions and effects, organize assets, and prepare exports and generation jobs.
Here's a short example using Codex:
https://www.youtube.com/watch?v=Owel8zkMWkY
For this demo I gave Codex almost no creative direction. I basically told it, you're live on YouTube, use Velorn to create a really cool motion graphics video. Just make something.
I wanted to see what would happen when the agent had access to an actual editing environment rather than asking it to execute a predetermined sequence of edits.
The video is about 3.5 minutes. I fast forward through some of the time where the agent is thinking, but otherwise it shows the interaction and the resulting work inside Velorn.
ComfyUI is not required for editing, captions, MCP, project management, or exporting. All generation in Velorn currently goes through a locally running ComfyUI instance on the same machine. For local models, your own GPU performs the work and Velorn doesn't charge credits. Generated results are brought back into the active project so they can immediately become part of the edit.
Some ComfyUI workflows can also use paid Partner Nodes/API services. Those may perform the actual computation in the cloud and require Comfy.org credentials and credits, but the workflow is still submitted through the user's local ComfyUI. Velorn doesn't currently operate a separate cloud generation backend. So someone can download Velorn and use it as an editor with MCP without configuring ComfyUI at all. ComfyUI is only required when using Velorn's current generation features.
Velorn uses Electron, React, Zustand for project/timeline state, FFmpeg for media processing and export, local Whisper for transcription, ComfyUI's API for generation, and a local MCP server exposing structured editorial operations.
I chose Electron because it gave me one cross platform editor implementation while still allowing integration with the filesystem, FFmpeg, local processes, ComfyUI, and MCP.
The project has picked up 400+ GitHub stars so far as a solo developer, I'm particularly interested in hearing from people who actually edit video or are experimenting with agents.
I would really love some feedback.
Velorn is free and open source. Try it!
To see longer MCP interactions without the agent's thinking being fast forwarded, I also recorded two uncut examples:
https://www.youtube.com/watch?v=_r4jf7ZDT2o
https://www.youtube.com/watch?v=AT9usQS3m48
Here's a music video that I made in just 2 days with Velorn that has gotten some attention:
https://www.youtube.com/watch?v=iX-YdjVMDhg
GitHub: https://github.com/VelornLabs/velorn\nWebsite: https://velorn.ai
Thanks for looking!", "5": "2026-08-22T18:05:29.494220"} +{"0": 147, "1": "hackernews", "2": "https://flownie.com/", "3": "Flownie \u2013 Open and Visual Data Workflow Platform with AI Agent Assistance", "4": "Flownie \u2013 Open and Visual Data Workflow Platform with AI Agent Assistance. ", "5": "2026-08-22T18:05:29.500173"} +{"0": 148, "1": "hackernews", "2": "https://fn2.ai/claude", "3": "Show HN: I built a Claude Code plugin to query 10.6M earnings-call embeddings", "4": "Show HN: I built a Claude Code plugin to query 10.6M earnings-call embeddings. Hi HN,
I've been using LLMs for one-off market questions against a database of earnings transcripts since ~2025. In tech years that's a long time so as a Solo dev I built a platform around it and have been adding miscellaneous other financial tools and integrations. Backend infrastructure is super boring! PostgreSQL & ElasticSearch, with the data appropriately sharded and replicated across two regions.
Some example requests:
"What did NVIDIA say about data-center demand on its last call?"
"Brief me on NVDA, AMD and AVGO before each market open"
"Notify me when AAPL moves more than 5% in a day."
The plugin is dependency-free Python and MIT licensed. The corpus, vector search and scheduled-agent service are hosted, so using them requires a free FN2 account. I\u2019m particularly interested in whether people find the scheduled-research part useful, but the basic features are free with integrations for Claude Code, Hermes Agent, OpenClaw, etc. Thanks!", "5": "2026-08-22T18:05:29.505758"} +{"0": 149, "1": "hackernews", "2": "https://news.ycombinator.com/item?id=49276601", "3": "Show HN: A better, free client for iMessage power-users", "4": "Show HN: A better, free client for iMessage power-users. TLDR: Your iMessage inbox is a mess. Go to attention dot righthand dot ai.
it's a free app for mac. It's like Superhuman for iMessage.\n___
You text your friends and family in iMessage. You text your colleagues in iMessage. If you run a business, you likely text prospects and leads in iMessage. I text our customers in iMessage - I want them to know they've got the fastest way to reach me. In 2026 you even text your AI agents in iMessage (Linq raises $20M to enable AI assistants to live in iMessage, Apple approves Poke as the first AI agent on its Messages for Business platform).
The point is: your iMessage inbox is overloaded. Even more than email, it's become the inbox where everything important happens. But it's not built for everything! It's basically optimized for friends & family groupchats.
Being an iffy texter is OK (debatable) with friends. But the founders and investors we talk to routinely describe this niche pain of co\u00f6rdinating business in iMessage. So we distilled the lessons from Superhuman, viz., the email experience should be fast af and beautiful and shortcut-laden, and we applied them directly to the inboxes that are hard to use - starting with iMessage.
Attention is the first app that allows you to become a power-user of iMessage.
I'm finally able to label and divide my iMessages into different folders. I can set reminders and follow-ups with a keystroke.
We are obsessed with the potential value of an agent who triages every message before you even see it. We love the concept of every new message arriving with pre-work done and a human-in-the-loop Approval Card curated by the agent with 3 differentiated next steps. It ships with an MCP - just point your favorite flavor of coding agent at it.
As far as the actual app: Your data is yours, fully local in LanceDB + SQLite & never leaves your machine unless you turn on the AI agent, at which point data just goes to whichever AI provider you chose.
For now, just know that Attention ships with AI features "off" by default and it will always be free to use.
attention dot righthand dot ai
this post was written without the use of language models by me, Joseph. would love to chat and answer any questions in the comments", "5": "2026-08-22T18:05:29.510875"} +{"0": 150, "1": "hackernews", "2": "https://federaide.rocklab.in", "3": "Show HN: FEDERaiDE, a TUI harness with P2P multi-agent routing and built in IDE", "4": "Show HN: FEDERaiDE, a TUI harness with P2P multi-agent routing and built in IDE. Hi!
Federaide is a general purpose multi-agent harness that runs in your terminal. It is meant for recreational programming and automating your scripts. The agents are just named instances of language models which have their own memories, backstories et cetera, and can coordinate with each other as they require. It has its own IDE (complete with interactive structure parsing and jump to def). \nI am building Federaide as a solo project.
Why do I build this? Some reasons:
- I reached out to some people to try my other IDE, many of them did not have suitable computers (or at all).
- During this time I observed almost every normal person has interacted with AI via their phones (usually the Gemini/ChatGPT app).
- I felt I could bring the joy of coding to a lot more people if I could make an AI assisted IDE run on Android.
- When I go to sleep I like to try out code ideas. Can't take my macbook to bed, have cats. Do have a spare Android, thought it would be nice to have a coding platform I could use in bed.
- Wanted to "vibecode" something, at the time gemini-cli was available, so this was the chosen project (tried a lot of concepts/variants, in rust go etc). Ultimately chose textual over a very simple reason: horizontal scrollbox (absolutely needed that for code display, may be OCD but linewrap does not do it for me).
- Wanted to create a truly powerful AI agent system without worrying about safety. Complete unabashed power was desired.
- Therefore termux native operation was a huge design goal.
- Kept wanting new features, kept adding them too.
- When I started this (in March) multi-agent harnesses were not a thing (arguably they still aren't mainstream, but there are other projects out there now). Wanted to see how different LLMs would react to each other in the same workspace.
- So here we are. It is at an early stage, there are often bugs, but if you find them, I will fix them. Thought I'd share it, it can be daily-driven at this point.
Does it solve a burning problem? Not really. Atleast not one other harnesses don't already (other than maybe providing a performant competent IDE for termux). Is it fun? Fuck yeah. Watch your agents delete the workspace while you sleep. And the excuses are gold :D
Well okay I do want to automate parts of wellsite data processing, the multimodal component of the active skills system is a step towards that, I mean what is geology if not hoarded knowledge and image processing on wetware? But the current models are so far not great at this stuff (I thought I would demo it, but so far the performance has not been up to my standards. Oh well will keep trying.)
This is completely free of MCP. Or any other borrowed compute as much as possible. Except the LLMs ofcourse, though I have tested with Llama.cpp. Everything that can be done on device will be done on device. No logins or email ids or anything like that (unless you count the ChatGPT subscription Oauth, which is not compulsory).
What do I use it for? I mostly use the research tool. Trying the paper trading (active skill) out has been fun (it was made on request for a friend). I like to experiment with orchestration, trying new ideas about memory and stuff like that, with the long term goal of making LLMs learn things outside of fine tuning or direct training. Running go/rust/python code directly on a phone has been fun.
Docs: https://federaide.rocklab.in/manual.html
Github: https://github.com/ROCK-LAB-PRIVATE-LIMITED/federaide
I have not put technical details here for brevity. Feel free to discuss them if you like.", "5": "2026-08-22T18:05:29.516763"} +{"0": 153, "1": "hackernews", "2": "https://imperavi.com/redactor/ai-assistant/", "3": "Show HN: Editor UI toolkit for building AI writing workflows", "4": "Show HN: Editor UI toolkit for building AI writing workflows. ", "5": "2026-08-22T18:05:30.932640"} +{"0": 156, "1": "hackernews", "2": "https://news.ycombinator.com/item?id=49329291", "3": "Ask HN: \u20ac300/mo to capture work you're doing. Tips how to reach right people?", "4": "Ask HN: \u20ac300/mo to capture work you're doing. Tips how to reach right people?. Hey Everyone,
My name is Salan and I'm a Researchrer at https://pdoom.org/ an early-stage AI Research Lab building LLMs for long-horizon tasks in Europe. We were funded by the German Government (SPRIND).
Currently, we are trying to acquire data, specifically from people who do open-source work/research. Therefore, we pay people \u20ac300 per month to capture their digital workflows.
We also built a custom tool that is privacy-preserving specifically for capturing these workflows: Get paid to record your work \u00b7 p(doom)
Since we have a high quality bar and need really strong people in their fields, finding participants hasn\u2019t been easy. Do you guys have any tips or ideas on how we can successfully reach these profiles?
Any help is highly appreciated!", "5": "2026-08-22T18:05:30.945618"} +{"0": 157, "1": "hackernews", "2": "https://www.moonlight.ng/captain/", "3": "Show HN: Captain, AI Travel Agent", "4": "Show HN: Captain, AI Travel Agent. I\u2019m a software designer from Lagos exploring conversational interfaces and frameworks for building agents. I made Captain to help with everyday travel planning. It can explore an itinerary, check live flight prices, or watch for the right time to book. Captain is a Telegram bot powered by a generalist model (anthropic/claude-sonnet-5), with specialist models handling workflows such as trip interpretation and voice transcription. It also includes a visual workspace where users can review the agent\u2019s work and directly manipulate their travel information. I'm building the agent to explore how conversation, tools, durable workflows, and direct manipulation can come together in one coherent product.", "5": "2026-08-22T18:05:30.951951"} +{"0": 158, "1": "hackernews", "2": "https://blaze.money", "3": "Show HN: Agents that run your finances", "4": "Show HN: Agents that run your finances. Hey HN. My name is Faiyam, cofounder/CEO of Blaze Money (YC S24). We're sharing here something new we've been working on in the agentic finance space for businesses. TL;DR -- you can use Blaze Money to run your business's finances agentically.
'Twas a long journey to get here. In 2023 we started in cross border payments after nomading around the world and feeling the struggle of paying international contacts. In the time since we started though the world has completely changed and now it seems obvious we want agentic money the way we want agentic everything else.
We've built an agentic finance offering for businesses. Our vision is to do both, something for individuals to manage their finances agentically and also something for businesses. We're getting started with businesses first.
We were inspired to build this for multiple reasons. One is, like I mentioned earlier, we were already in the money space, so we were thinking about products in that space as is. At the same time we were building so much internal AI tooling for our own dev and business workflows that we had become completely accustomed to just doing everything through Claude (our harness of choice). It almost feels archaic to now need to open up a banking app to send money or use an app to request funds from someone. We wanted to just be able to text an agent "hey send vendor X the funds for their latest invoice," and be able to get that done.
Also, another thing that kept coming up is that we're a super lean team, and I in particular would spend a lot of time, and more importantly mental energy, each month just trying to manage our finances. It was an always pervasive anxiety that I had to deal with which would slow me down. I want to be able to just chat with an AI about our finances to get clarity, make plans, do payments, get all that stuff off my head.
We made it accessible a bunch of different ways. You can install it via npx @blaze-money/cli and use it via Claude Code, Codex, etc., you can tell your Hermes/openclaw to "install npx @blaze-money/cli" and get me setup, or you can use it in a good 'ol fashioned UI (so 2023, I know) on our site if you want as well.
Here are some of the things you can do. Once you've connected all your accounts, you can do a lot of the finance and analysis work your fractional CFO would otherwise do, like forecasting or runway planning. You can do preaccounting to help your accountant get things done faster at month's end. And since we started in payments, we've built that in so you can send and receive money globally via a bunch of rails, some traditional some using stablecoins, to settle all your finances and not just analyze them.
I'm really proud of what we built because I already use this regularly and get value from it, we have friends and family that are already using this, and we want to now share this with a wider community to get more eyes, ears, feedback, whatever you have to share.
Thanks for taking the time to take a look at what we've built and please feel free to give us your honest feedback, suggestions, feature requests, or philosophical pontifications on the nature of money in a post digital superintelligence world...", "5": "2026-08-22T18:05:30.956767"} +{"0": 159, "1": "hackernews", "2": "https://github.com/Winipedia/pyrig", "3": "Show HN: Pyrig \u2013 A tool that automates project setup and maintenance", "4": "Show HN: Pyrig \u2013 A tool that automates project setup and maintenance. # What is pyrig?
pyrig is a package and tool that *rigs up* your project. It scaffolds and initializes a complete, fully configured, installed and working Python project with one command and makes the process of developing and maintaining it more seamless and efficient by automating things like configuration management, CLI generation, testing infrastructure, and more.
# Requirements
* Python 3.12+\n* Git\n* uv
# Quick Start
uv init my-project --python 3.12\n cd my-project\n uv add pyrig --dev\n uv run pyrig init\n# Features
# [Project Scaffolding & Initialization](https://Winipedia.github.io/pyrig/scaffolding)
`pyrig init` generates a complete project in one command that works out of the box. This includes everything a modern python project needs:
* Standardized directory structure\n* Fully configured dev tools (linters, formatters, type checkers, test frameworks, git hooks, etc.)\n* End-to-end CI/CD pipeline with GitHub Actions and integrated repository protection\n* Complete and working CLI\n* And much more...
# [File & Configuration Management](https://Winipedia.github.io/pyrig/config-files)
Every generated file is backed by a Python class that validates and merges automatically. Override any config by subclassing, or define entirely new config files \u2014 pyrig discovers and manages them for you. Run `pyrig sync` to create or update all config files at once. Run `pyrig mk subcls` to generate a subclass for overriding a specific file.
# [Automatic CLI](https://Winipedia.github.io/pyrig/cli)
`pyrig init` sets up a CLI for your project that works immediately. Generate and add new commands by running `pyrig mk cmd <name>`. An automatic version command is included that shows the version of your project. Run `my-project version` to see it in action.
# [Mirror Test Generation & Maintenance](https://Winipedia.github.io/pyrig/mirror-tests)
Generate test skeletons with `pyrig sync`. This will generate test skeletons for all source modules and update them automatically as your project evolves.
# [Multi-Package Inheritance and Extensibility Architecture](https://Winipedia.github.io/pyrig/architecture)
Override and customize any and all behaviour to suit your project's needs. pyrig's classes are designed for inheritance and composition, allowing you to create custom configurations, tools, and more by subclassing and simply overriding methods. pyrig will automatically discover and use your custom classes without any additional configuration. Run `pyrig mk subcls` to generate a subclass for any pyrig class.
# [CI/CD & Repository Protection](https://Winipedia.github.io/pyrig/ci-cd)
Pyrig generates GitHub Actions workflows for CI/CD which automatically test and release your code. They also configure and apply repository protection settings and protection rulesets. Push your code to GitHub after initialization and see it in action.
# Commands
Run `pyrig --help` to see a list of all available commands and their usage. Run `pyrig <command> --help` for more information about a specific command and its usage. Run `my-project --help` to see the automatically generated CLI for your project.
# Documentation
|[*Full Documentation*](https://Winipedia.github.io/pyrig)|The manually written documentation|\n|:-|:-|\n|[*CodeWiki*](https://codewiki.google/github.com/winipedia/pyrig)|AI-generated documentation|\n|[*Tutorials*](https://www.youtube.com/@Winipedia-py/playlists)|YouTube tutorials for pyrig|", "5": "2026-08-22T18:05:30.962640"} +{"0": 160, "1": "hackernews", "2": "https://github.com/armin5872/OptiQra", "3": "Show HN: OptiQra \u2013 Open-source AI website optimization and intelligence platform", "4": "Show HN: OptiQra \u2013 Open-source AI website optimization and intelligence platform. Hi HN,
I've been building OptiQra for quite a while, and I finally finished the desktop version.
OptiQra is an open-source website intelligence and optimization platform. It can crawl and analyze websites or local projects, identify SEO/AEO/GEO, accessibility, performance and technical issues, explain them with AI, and automatically fix them.
The desktop version is where the project gets particularly interesting. It can run locally, work with local projects, operate offline for analysis that doesn't require cloud services, and run scheduled scans in the background.
The workflow is roughly:
URL/project \u2192 crawl \u2192 analyze \u2192 understand the issues \u2192 inspect them on the actual site \u2192 fix them \u2192 verify the result.
It also has a fairly visual crawler, including live 2D and 3D crawl trees, and the AI interface is customizable rather than being limited to a single assistant personality.
I built the UI and CSS myself rather than using Tailwind/shadcn, and the project has grown considerably beyond a simple SEO checker. There are currently 150+ analysis rules along with integrations and additional tooling.
This is the first release where I feel the desktop application is actually ready for people other than me to use.
GitHub: https://github.com/armin5872/OptiQra
website : https://optiqra.vercel.app
I'd especially appreciate feedback from people who work on large websites, developer tooling, SEO, or web infrastructure.
I'm also interested in hearing what you think is missing or what you'd remove. The project is large enough now that outside perspective is probably more valuable to me than another feature I can think of myself.", "5": "2026-08-22T18:05:30.970751"} +{"0": 161, "1": "hackernews", "2": "https://argonix.io/", "3": "Show HN: Argonix \u2013 One AI-powered platform for SRE and cloud operations", "4": "Show HN: Argonix \u2013 One AI-powered platform for SRE and cloud operations. Hi everybody,
I built https://argonix.io.
Argonix is an AI-powered platform for SRE, security and cloud operations.
The idea is to bring and connect together things that are usually spread across many different tools: monitoring, security, Finops, health checks, workflows and incident investigation.
A few things you can do with it:
Connectors: Connect Argonix to your existing stack: Kubernetes, AWS, GCP, Azure, your own MCP servers, etc.
Workflows: Build automation flows using the connectors.
Health notebooks: Get a daily overview of the health of your platforms. Argonix can publish the report to Slack, Confluence, etc., and create tickets when something needs attention.
Periodic jobs: Schedule recurring checks, analysis and automation.
Security: Scan of your connectors infras, your code also (new feature), analyse network logs, and look for vulnerabilities and threats. It can also consume data from tools such as Falco, Wazuh, Wiz, SonarQube, etc.\nFinOps \u2013 Analyse cloud costs across AWS, GCP and \nAzure.
Monitoring: Monitor works just like UptimeKuma or others, plus automatically discover and monitor Kubernetes ingresses/gateways, run synthetic tests, and analyse logs.
Alert investigatio: You can send alerts from Datadog, Grafana and others to Argonix. The AI can investigate the alert, read the KB and provide recommendations or even propose a PR.
Conversational AI: Talk to your infrastructure and connected tools through a conversational interface.
Under the hood, Argonix is built with Python/DRF, Vue.js and PostgreSQL.
Argonix is available as both a SaaS and a self-hosted version.
You also choose which AI model you want to use. You can connect your own AI provider/account, and with the self-hosted version you can run a local model. I've been testing it locally with Qwen.
You can create an account for free and start playing with it.
It's still a work in progress and there are definitely some bugs, so I'd really appreciate feedback from people working in the field.
I would love to hear what you think, what you would change, and what would make it genuinely useful in your day-to-day work.
Thanks,\nDavid.", "5": "2026-08-22T18:05:30.975882"} +{"0": 162, "1": "hackernews", "2": "https://www.salestrics.com/", "3": "Show HN: Salestrics \u2013 An open MCP server and CRM for AI-native revenue teams", "4": "Show HN: Salestrics \u2013 An open MCP server and CRM for AI-native revenue teams. Hey HN, I\u2019m Austin, founder of Salestrics (salestrics.com).
The Problem: > Most AI agents today are trapped in chat windows. While models are smart enough to run complex multi-step workflows, giving them production access to business context (CRM, support, billing) usually means managing unsafe local API keys or hacking together fragile point-solution scripts.
What We Built:\nWe built Salestrics to serve as both an all-in-one revenue workspace (CRM, service desk, billing, docs, email) and a production-grade Model Context Protocol (MCP) execution proxy.
How It Works:
Unified Context: Instead of fragmenting data across five SaaS tools, customer records, tickets, and invoices live in a single data layer.
165-Tool MCP Server: You connect your local AI environment (Cursor, Claude Desktop, local LLMs) once via our MCP proxy.
Full-CRUD Execution: Your agent gets zero-config execution access across native apps and third-party tools (Stripe, PostHog, Sentry)\u2014allowing it to do things like resolve support tickets, issue refunds, or update pipeline stages directly from your IDE or chat client.
Human-in-the-Loop Governance: High-impact agent actions (deleting records, mass messaging, issuing payouts) trigger explicit approval gates and immutable audit logs before execution.
Traction & Tech Stack:\nWe launched two months ago and currently power 320+ active organizations. The backend is built with high-throughput node/TypeScript orchestration, connected to a dual-model AI routing engine (Salestrics-AI-v1/v2).
Try It Out:\nWe have a Free Forever tier for solo builders. You can grab your MCP keys and test the proxy immediately without putting down a credit card.
I\u2019d love to hear your thoughts on our MCP architecture, human-in-the-loop security patterns, or what tools/actions you'd want added to the MCP server next!", "5": "2026-08-22T18:05:30.981631"} +{"0": 163, "1": "hackernews", "2": "https://www.physicsbase.ai/", "3": "Show HN: Physicsbase.ai Validated Physics for AI", "4": "Show HN: Physicsbase.ai Validated Physics for AI. HI hackernews community,
I introduce physicsbase, which is a tool that allows ai to carry out numerical simulation of phyiscs problems using Finite element analysis (FEA). The AI agent can decide to create the geometry and submit it for analysis or it can send an input geometry, meshed or unmeshed CAD file and physicsbase will return a validated FEA simulation result, the ai can decide to iterate on the model to reach a design objective or decide to present the model to the user.
With the current proliferation of AI CAD startups, this tool will enable them to bring physics simulation into their workflow easily.
I will love to hear what you guys think.", "5": "2026-08-22T18:05:30.989567"} +{"0": 164, "1": "hackernews", "2": "https://github.com/jajego/interlock", "3": "Show HN: Interlock \u2013 Atomic TypeScript domain transitions on PostgreSQL", "4": "Show HN: Interlock \u2013 Atomic TypeScript domain transitions on PostgreSQL. Hello HN! I built Interlock because I kept seeing the same architectural issue arise in backend code at work, where a state change starts as one update, then slowly accumulates features until the transaction logic is spread everywhere. This has been exacerbated in the AI era of hyper iterative development, and it confuses devs and agents alike.
Interlock is my attempt to make that transition a single explicit unit and commit all of its durable effects together in PostgreSQL.
PostgreSQL is the only supported backend right now. I was torn about some of the packaging stuff, especially the separate core and Postgres packages.
The guarantees, reference app, concurrency tests, failure cases, and implementation details are all in the repo. There are plenty of powerful tools for modeling state and running workflows; Interlock is aimed at the narrower case where you want something lightweight for atomic, database-backed transitions.
I\u2019d be interested to hear whether the abstraction feels useful, where the boundaries seem wrong, or stuff I\u2019ve overlooked.
Thank you!", "5": "2026-08-22T18:05:30.993524"} +{"0": 165, "1": "hackernews", "2": "https://slickfast.com/", "3": "Show HN: SlickFast Deterministic Chart/Dash Renderer, No Browser(JSON \u2192 SVG/PNG)", "4": "Show HN: SlickFast Deterministic Chart/Dash Renderer, No Browser(JSON \u2192 SVG/PNG). SlickFast started 6 months ago, and evolved in a super backwards way. I was using lowfruits to look for good SEO keywords, I wanted to make a simple free tool to rank with SEO. I found some great keywords related to graphs/charts. I did research for optimizing for SEO. Turns out edge processing / using static HTML is super fast and lightweight, great for SEO. I made freepiechartmaker.com. I was really blown away by how fast the site was loading, and how lightweight all the processing was. The site renders changes on the fly, and is much much faster than other sites in the space. I started looking into the tech, pure math rendering, and I saw a lot of openings for what this tech can do. That's how SlickFast was made.
SlickFast is a JSON input > pure javascript SVG Native render core > with PNG+SVG output. No headless chrome. No library calls. deterministic output. on my local machine(m1 max) it renders 140,000 svg charts a second. PNG @ ~50/sec at retina (scale-2) and ~145/sec at scale-1, single-core. (we have a benchmark mjs in our github repo)
SlickFast can render 47 chart types. Any resolution, aspect ratio, any tile combination, easily deliverable anywhere, and made for agentic workflows. \nOur MCP is free and open source.
Tiling allows complex dashboard creation. Each dashboard is only one render. SlickFast API serves URL endpoints. SlickFast is lightweight, we have a hero dashboard demo at slickfast.com/deck that is updating every 3 seconds. This is millions of renders, and the cost per month is less than $10.
SlickFast was intended to be ready for agentic workflows, and visualizing agentic output as cheap and frictionless as possible.
A benefit to the SlickFast deterministic pure math render core, is that edge servers will easily cache images. Millions of people can see a SlickFast endpoint, and only a few renders will actually be used, because the edge servers will happily cache our very small charts.
SVG Native chart rendering provides for retina quality, at 50kb image sizes.
Our favorite demo is LIVE GitHub dashboards, living right on the repo page. just put a SlickFast URL in your readme. You can see our living dashboard at github.com/slickfast/slickfast. We have made githubs templates for the public. you can easily add your own dashboard. https://github.com/SlickFast/github-dashboard-template
SlickFast also has telegram deliverable templates. https://github.com/SlickFast/telegram-weather-template - if you go to the repo, the weather dash being sent via telegram, is on the repo as a live dashboard - Its updated everyday with a real weather report.
Another great demo is our X.com NYC air quality index bot (@NYCAIRREPORT). Slickfast posts a AQI dash on X.com 3 times a day. With the new x API terms, each post is only about 2 cents. The cost on the slickfast side is basically nothing. Our free API tier currently provides 250 renders a month. Enough space for a few bots. Because slickfast was made for agentic use, an agentic was able to design the entire dash and inputs in only a few takes. AI Agents, can nearly one shot most tasks. One or two minor revisions is usually all it takes.
Slickfast.com Github.com/slickfast/slickfast
Slickfast is free and open source under AGPL license. I'm so delighted to provide some real value for free. We have a roadmap with some super cool things coming. Slickfast is very powerful for email, or tying to APIs.
Thanks Everyone! I would love to hear your thoughts.", "5": "2026-08-22T18:05:30.999821"} +{"0": 166, "1": "hackernews", "2": "https://edotenv.com/", "3": "Launch HN: EdotEnv (YC S26) \u2013 Quant Trading RL Envs to Teach LLMs Research", "4": "Launch HN: EdotEnv (YC S26) \u2013 Quant Trading RL Envs to Teach LLMs Research. We are Rui and Michael and we\u2019re building EdotEnv (https://edotenv.com): self-improving RL environments from Quant Trading workflows.
With all the benchmaxxing around, evals saturate and become meaningless for model comparison. Useful benchmarks should increase in difficulty as models advance. Back in our Quant jobs, Michael and I saw that the market has exactly this property: markets became more efficient as people profited from trading inefficiencies, making new profitable strategies harder to find and old ones decay over time.
This makes markets an ideal, continuously evolving benchmark for LLM training. The hard part is to turn professional quant workflows into reliable training envs, as this is a very niche expertise.
In our environments, we give LLMs a quant trading workflow and evaluate their performance on out-of-sample data: build predictive features/ models, design a portfolio, backtest strategies, adapt continuously to market regimes. Each step is a task with different self-built tools. For example, a predictive feature building task gives the agent cleaned market data of time period [0,T] to research ideas, a backtesting tool to test created features at time t on [0, t], an execution tool to trade strategies with the new features on [t+1, T] and a final evaluation. Our reward isolates the agent's feature building skills and yet benefits from market properties.
From running SOTA models in our environments, we see that i) they seem to struggle with iterating deeply on research ideas, preferring broad shallow searches; ii) higher reasoning does not seem to increase performance and iii) agents do not understand trading, e.g. when losing money they stop trading instead of trading smarter. Check out our blogs for more details! https://edotenv.com/?tab=blog
Quant workflows are essentially applied ML research, long-horizon planning and continual learning. Through our envs, we teach these transferable research skills, rather than task specific answers. Our environments are closer to a realistic research workflow: we use real-world data instead of synthetic ones; our envs naturally contain noise and real trade-offs; our rewards are verifiable and immediate, with no need for an additional LLM judge or human expert.
We open sourced a sample task repository: https://github.com/MMcollab-dotcom/feature-engineering. We plan to sell continuously improving envs to AI labs/researchers/enterprises training their own agents, who are interested in ML modelling capabilities, continual learning, long horizon planning or Quant Research in general.
We'd love feedback from anyone trying out their own agents in our envs, for either eval or post training. And of course, we are always happy to discuss the future of trading with LLMs (and no, it should not be asking the LLM to read tea leaves and give you the stock to buy tomorrow). Looking forward to your comments!", "5": "2026-08-22T18:05:31.005768"} +{"0": 167, "1": "hackernews", "2": "https://github.com/RudderCode/Rudder", "3": "Show HN: I Repurposed Unit Tests to Show How Much Coding Agents \"Improvise\"", "4": "Show HN: I Repurposed Unit Tests to Show How Much Coding Agents \"Improvise\". As someone who is using agents to code for me a lot lately, I was starting to feel more and more like I'm losing touch with the code I write--it's starting to feel like Claude's code rather than mine. I'm pretty sure I'm not alone in feeling this way, so I made Rudder as a first earnest shot at trying to fix this problem.
Nowadays with AI coding agents, unit tests are pretty much dead code--your agent may write them because it has to hit coverage, but it is basically meaningless because it's just checking against its own implementation. So, I figured I would repurpose unit tests back to their actual core purpose: making sure that your code does what you want it to do.
Rudder is a local Codex/Claude Code plugin that captures your coding session history and forces your agent to rewrite unit tests exclusively from the intent expressed in your prompts, turning your unit test coverage into a proxy for what percent of the code actually comes from decisions you've made talking to your agent. After showing you an initial coverage percentage, Rudder prompts you with targeted questions to increase coverage to your goal target, and runs a red-green test-driven-development flow to get you up to your goal. Hitting coverage is the equivalent of saying "My decisions in chat conversation cover X% of the code here."
There's a few other ideas out there for getting agents to code closer to the user's intent including some spec-driven-development tools, etc. But I was pretty unsatisfied with the idea that I'd have to completely change my workflow to be able to make decisions when coding with AI. So, I tried to make Rudder as unintrusive as possible by embedding directly into your existing coding sessions.
Install the plugin into every coding agent you want to track and it will work cross-platform (i.e. if you code in the same worktree with both Codex and Claude Code, Rudder will be able to consolidate the prompts from both by tracking the working branch). When you want to use it, just tell your agent "Run Rudder" and it will kick off the flow.
It's Apache-2.0 licensed, because that seems popular nowadays. I'm actively improving this day-by-day for my own use, so let me know what you think! Especially curious how you guys go about making your agent's code your own and verifying that your actual intent makes it into the code your agents generate.
Check it out here: https://github.com/RudderCode/Rudder", "5": "2026-08-22T18:05:31.013687"} +{"0": 168, "1": "hackernews", "2": "https://github.com/maxrodrigo/clai", "3": "Show HN: Clai \u2013 AI for the command line (stdin \u2192 LLM \u2192 stdout)", "4": "Show HN: Clai \u2013 AI for the command line (stdin \u2192 LLM \u2192 stdout). Since ChatGPT launched I've been trying to integrate LLMs into my workflows; emails, commits PRSs, or even long chat threads.\nMy instinct was to code a small automation platform, but I found myself limited once again by its own constraints. Went back to zero, twice, and went back to the tools I've been using for years: UNIX pipelines.\nI started using llm and later mods (now deprecated) and they're great tools, but I wanted something leaner and that would compose better with other tools.
Clai does one thing: takes stdin, sends it to a model of your choice and prints the result. No REPL, no state. When it's done it just exits.
git diff | clai commit\n cat article.txt | clai summarize | glow\n pbpaste | clai tldr\n curl -s example.com/article.html | clai -e "Extract the three main concepts"\n\nWorks with major providers and I'll keep adding more. It can also be pointed to a local model so nothing leaves your machine.
Install with:
brew install maxrodrigo/tap/clai\nI\u2019m Dallin, building Keystroke with my co-founder Blake. Keystroke is a recent pivot from Buster, our original YC W24 company.
Keystroke lets you build agents and durable workflows as TypeScript in your own repo, then deploy them to a web application where your team can use them. Our repo (ELv2 licensed) is here: https://github.com/keystrokehq/keystroke.
At Buster, we spent a lot of time building internal agents and automations. We used visual platforms like n8n and Zapier, as well as code-first tools like Trigger.dev, Mastra, and several of Vercel\u2019s open frameworks. We genuinely liked many of them, but kept encountering two different problems:
1. Visual platforms gave us integrations, credentials, hosting, and an app the rest of the team could use, but became difficult to test, review, and debug as our systems grew.
2. Code-first tools gave us the flexibility of ordinary TypeScript and worked well with coding agents, but left us building the UI, runtime, permissions, credentials, and operational tooling ourselves.
So, we started building the thing we wish existed: a code-first framework that made it easy to build complex agents and AI systems, with a web app for teams to collaborate on top. We liked it so much that we decided to pivot away from Buster and focus on building Keystroke full time.
With Keystroke, I was able to rebuild our W24 idea (an AI data analyst) in ~5 minutes. I pointed Cursor at our old Buster repo and it was able to:
- spin up a local Keystroke project
- rebuild the OG Buster agent w/ memory, its own file system, etc
- connect our Postgres DB, PostHog, Metabase, and Slack
- build skills for using each of the data sources
- explore our DB and add schema docs to the file system
- build a workflow to keep schema docs synced
- name the agent "Delbert"
- then pushed him up to the Keystroke app
Since then, \u201cDelbert\u201d has been answering ad-hoc questions for me in Slack on a regular basis. He builds me metabase dashboards when I need them. He sets his own triggers, reviews key metrics periodically, and pings me in Slack if anything looks off.
In Keystroke, workflows are ordinary async TypeScript: a `for` loop is an actual `for` loop, inputs use Zod schemas, and everything you build can be tested with Vitest. Our managed runtime makes workflows durable across errors, sleeps, approval steps, and process restarts.
Agents wrap Vercel\u2019s AI SDK and can use workflows, actions, other agents, MCP servers, and integrations as tools. They also include memory, a persistent filesystem, web search, triggers, and optional VM sandboxes.
Deployed agents and workflows instantly appear in the web app with a chat UI, workflow visualizations, input forms, sharing, credential management, integrations, and logs.
The entire platform was built from the ground up with coding agents in mind. Your coding agent can use the CLI to scaffold a local directory with a Keystroke project and an AGENTS.md. Then, it can build whatever agent or AI system you want, search our docs, run tests, and push what you build up to the Keystroke platform.
We plan to make money from our cloud offering, which has a free plan and usage-based pricing.
Our docs are here: https://keystroke.ai/docs.
You can watch a quick demo of Keystroke here: https://supercut.ai/share/keystroke/zeor1pD0bNlSTOuLUWmRBT?v....
You can try on cloud for free at: https://keystroke.ai/.
We\u2019d especially appreciate feedback from people who have maintained agents and AI systems in visual tools or code-first frameworks. Where does this approach seem useful, and where are we rebuilding something that already works well elsewhere?", "5": "2026-08-22T18:05:31.025721"} +{"0": 170, "1": "hackernews", "2": "https://hoplite.sh", "3": "Launch HN: Hoplite (YC S26) \u2013 Effortlessly deploy cloud coding agents", "4": "Launch HN: Hoplite (YC S26) \u2013 Effortlessly deploy cloud coding agents. Hi HN, we\u2019re Bence and Ryan, founders of Hoplite (https://hoplite.sh). Hoplite lets you deploy coding agents in the cloud, with a suite of tools that makes it incredibly easy to QA features. During onboarding, we port over your local setup - sessions, memories, MCP servers, and get your projects ready to run in the cloud.
Here\u2019s a demo: https://youtu.be/bnyktZ_9pjE
We got here after pivoting away from the idea we applied to YC with; AI for retail investing. It ultimately wasn\u2019t a product that we ourselves would use, nor served a customer base that we felt connected to. In reflecting on what we really wanted to do, we realised that we loved talking to founders and developers, and were really opinionated about the specific area of cloud agents. We tried out all the existing solutions, and didn\u2019t find one that A) took good advantage of being in the cloud, and B) was performant and felt good to use.
We\u2019re building a product that we feel reflects what mainstream development will look like in 6-12 months. As models improve, developers will end up reviewing less and less code, and will instead focus on reviewing the product output. That means evaluating new user flows, visually verifying that new features look good, that the API works as expected, that the CLI works on Windows, etc. And doing it while running hundreds of agents concurrently.
On the agent side, we\u2019ve created a custom harness. We spent a lot of time deciding on whether we should use an off the shelf solution like Codex/Claude Code, but ultimately wanted the independence and freedom that came with building it in house. It also means that we can test out completely new features without relying on Anthropic and OpenAI to catch up.
Everything is hosted on AWS, with the exception of: Temporal for durable workflows, Modal for sandboxes, and Planetscale for our database. Our infra decisions were driven by a strong belief that agents are becoming a tier 0 piece of infrastructure, and they need the reliability and security to match that.
You can try it now for free with the code \u2018HACKERNEWS\u2019 - we\u2019ve included $100 in free credits, plus you can connect your Codex subscription and use OpenAI models via it. You can see some more details around our pricing at https://hoplite.sh/pricing.
At the moment we\u2019re focusing on optimising two key experiences: onboarding and previews, and would love to hear your feedback on them. And if you find that the agent's performance in certain tasks doesn\u2019t match your expectations, please let us know!", "5": "2026-08-22T18:05:31.032137"} +{"0": 216, "1": "producthunt", "2": "https://www.producthunt.com/products/bitdrift", "3": "bitdrift.ai", "4": "bitdrift.ai.
\n The world\u2019s first agentic mobile observability platform\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-23T07:00:02.339533"} +{"0": 217, "1": "producthunt", "2": "https://www.producthunt.com/products/protonote", "3": "ProtoNote", "4": "ProtoNote.\n Share AI-built prototypes, get feedback pinned to the page\n
\n\n Discussion\n |\n Link\n
", "5": "2026-08-23T07:00:02.355533"} +{"0": 221, "1": "hackernews", "2": "https://github.com/gojiplus/layoutlens", "3": "Show HN: LayoutLens: AI-Powered Visual UI Testing", "4": "Show HN: LayoutLens: AI-Powered Visual UI Testing. ", "5": "2026-08-23T07:00:09.806651"} +{"0": 222, "1": "hackernews", "2": "https://news.ycombinator.com/item?id=49405405", "3": "Show HN: Hands-Rust MCP/CLI that sees the Windows desktop and clicks real Chrome", "4": "Show HN: Hands-Rust MCP/CLI that sees the Windows desktop and clicks real Chrome. I built Hands because I wanted a coding agent to use this Windows PC and a real Chrome profile the way I do: look at the screen, move the real mouse, type, click , without turning Chrome into an automation browser.It is a Rust MCP/CLI. A harness (Grok, Codex, Claude Code, OpenCode, etc.) calls tools like observe, click, type, scroll. Observe is a screenshot path plus a small element list (UIA + optional Chrome DOM ids). Click is OS SendInput on a B\u00e9zier path, not a Chrome DevTools click.
There is no Playwright, no Puppeteer, no remote debugging port. Daily Chrome is launched with no extra flags, or attached if it\u2019s already open. Sites that key on CDP/automation flags mostly don\u2019t see that. They can still see injected input (LLMHF_INJECTED).
A tiny unpacked Chrome extension can fuse page structure (chr: ids, listing cards) so the model isn\u2019t guessing from pixels. Sideload is manual. Fusion dies if the service worker goes inactive; reload the card.
What it is good for: personal research on your own desk. \u201cFind a Camry on cars.com,\u201d read a page, fill a ZIP, dismiss a cookie banner.
What it is not:\n\u2022 Not a sandbox. It can click whatever is on screen, including checkout and Easy Apply.\n\u2022 Confirm-before-money is best-effort classification in the binary, not a guarantee. Prompt injection from the screenshot/DOM is real; the binary treats that text as untrusted, the model might not.\n\u2022 Not a CAPTCHA solver on daily Chrome. Two visible tries, then it yields and waits for the puzzle to go away.\n\u2022 Windows only. \n\u2022 Install is: build the exe, register a native-messaging host, sideload the extension, point an MCP client at hands mcp. README is the runbook. Missing an API key does not fail the build; do_task is optional.\n\u2022 Logs live under %LOCALAPPDATA%\\hands\\logs\\. The extension asks for <all_urls> so it can map the tab you\u2019re looking at.
Repo: https://github.com/Ryan-AI-Studios/hands (MIT)
Happy to answer how observe/fusion/the fence work. If you try it, Pause/Break is the kill switch.", "5": "2026-08-23T07:00:09.814876"} +{"0": 223, "1": "hackernews", "2": "https://agent2creator.vidmoat.com", "3": "Show HN: Agent2Creator \u2013 a video social network whose members are AI agents", "4": "Show HN: Agent2Creator \u2013 a video social network whose members are AI agents. ", "5": "2026-08-23T07:00:09.821417"} +{"0": 224, "1": "hackernews", "2": "https://meetless.ai", "3": "Show HN: Active Source of Truth for Your Coding Agents", "4": "Show HN: Active Source of Truth for Your Coding Agents. Howdy! Happy Saturday everyone!
As a solo founder, I have always tried to maximize my speed by letting coding agents build as much as possible in parallel. However, as an engineer, I don't trust that AI will always make the right decisions and work with the right context. In the past, I always needed to click through my sessions to glance at the AI's output, try to understand what it was doing, and hopefully steer it or stop it in time.
As a result, the maximum number of concurrent sessions I could manage at once was only 4. I didn't want to be the bottleneck, so I built Meetless Agent (MLA). It basically does what I had to do manually before:
- Monitors the coding agent's tasks and actions to supply it with the correct, up-to-date context.
- Continuously reconciles running information (such as provided/tagged documentation, the agent's output, and the agent's decisions) to actively maintain the source of truth at all times.
- Keeps track of the current rules for the repo, and if an action triggers a registered rule, the rule is injected into the agent context.
My benchmarks show that running coding agents with the help of an active monitor improves quality and accuracy, consumes fewer tokens, and finishes faster: https://research.meetless.ai/stale-context/
Of course, the agent alone can't decide the source of truth; it requires human review and decisions for contradictions, etc. But for the most part, it can safely build a consistent ontology of the current source of truth.
From this, I want to build an AI layer to maintain the source of truth across the business, so I will release more connectors for Slack, Jira, Confluence, etc., soon. The goal is for this AI to assist in every part of the business. Eventually, the same coordination layer will understand that a decision made in Slack affects a Jira task, a document, an email conversation, and what a coding agent should do next without every tool becoming another isolated memory silo.
The coding agent connector is open source at:\nhttps://github.com/Meetless/mla
I am looking forward to your feedback!", "5": "2026-08-23T07:00:09.825357"} +{"0": 225, "1": "hackernews", "2": "https://knowledgelens.ai/", "3": "Show HN: Lens AI \u2013 generate your JSON-LD files and keep them up-to-date", "4": "Show HN: Lens AI \u2013 generate your JSON-LD files and keep them up-to-date. ", "5": "2026-08-23T07:00:09.831652"} +{"0": 226, "1": "hackernews", "2": "https://www.macaiapps.com/", "3": "Show HN: MacAIApps \u2013 a directory of AI-powered Mac apps", "4": "Show HN: MacAIApps \u2013 a directory of AI-powered Mac apps. I love discovering Mac apps that use AI, so I built MacAIApps to collect them in one place.
You can browse by category and find apps you might otherwise miss.", "5": "2026-08-23T07:00:09.836530"} +{"0": 227, "1": "hackernews", "2": "https://spielplayer.com/", "3": "Show HN: Spiel \u2013 watch your own files on every device", "4": "Show HN: Spiel \u2013 watch your own files on every device. ", "5": "2026-08-23T07:00:09.841163"} +{"0": 228, "1": "hackernews", "2": "https://aurabid.now/", "3": "Show HN: Aurabid, pay more to get ranked #1 and beat others", "4": "Show HN: Aurabid, pay more to get ranked #1 and beat others. Indie hacker help needed!
My aim was to make aurabid for humans
i.e. me bidding $10
Feedback is showing that I didn't make it clear enough that it's for humans, not companies
Also built a leaderboard per country/city, so that you can claim #1 in diff places
How do I make it clearer that it's just for humans? Thanks for any help!", "5": "2026-08-23T07:00:09.845844"} +{"0": 229, "1": "hackernews", "2": "https://github.com/debabratasaha-dev/techskills", "3": "Show HN: TechSkills \u2013 Open-source skill modules for AI coding agents", "4": "Show HN: TechSkills \u2013 Open-source skill modules for AI coding agents. ", "5": "2026-08-23T07:00:09.855747"} +{"0": 230, "1": "hackernews", "2": "https://github.com/musoyangrigor/gitx-skill", "3": "Show HN: Git workflow as an AI-agent skill", "4": "Show HN: Git workflow as an AI-agent skill. ", "5": "2026-08-23T07:00:09.861699"} +{"0": 231, "1": "hackernews", "2": "https://github.com/musoyangrigor/scroll-video-website-skill", "3": "Show HN: I turned Apple-style scroll-video websites into a reusable AI skill", "4": "Show HN: I turned Apple-style scroll-video websites into a reusable AI skill. ", "5": "2026-08-23T07:00:09.867945"} +{"0": 232, "1": "hackernews", "2": "https://github.com/namo-robotics/namo_complete", "3": "Show HN: Namo_complete: a non-obtrusive AI autocomplete for the bash terminal", "4": "Show HN: Namo_complete: a non-obtrusive AI autocomplete for the bash terminal. I created this simple AI autocomplete tool for the bash terminal. I'm sure many like it exist already, but unlike the others this one does not completely hijack your terminal and stays mostly out of the way. Contributions/feedback are welcome. One other interesting thing about it is that it's written in a custom memory-safe (aspirationally) language that I am developing.", "5": "2026-08-23T07:00:09.873333"} +{"0": 233, "1": "hackernews", "2": "https://riyadhtechweek.com/", "3": "Show HN: Riyadh Tech Week \u2013 a year-round guide to Saudi Arabia's tech ecosystem", "4": "Show HN: Riyadh Tech Week \u2013 a year-round guide to Saudi Arabia's tech ecosystem. ", "5": "2026-08-23T07:00:09.878310"} +{"0": 234, "1": "hackernews", "2": "https://bookerapp.replit.app/book/ai-cognitive-gym/ai-as-cognitive-gym-equipment", "3": "Show HN: AI as Cognitive Gym Equipment", "4": "Show HN: AI as Cognitive Gym Equipment. Why AI assistance should be judged by what it does to your future thinking, not just your next answer", "5": "2026-08-23T07:00:09.882199"} +{"0": 235, "1": "hackernews", "2": "https://apps.apple.com/ie/app/clearvoice-text-to-speech/id6798899505", "3": "Show HN: ClearVoice TTS, SOTA voice cloning model running offline, in an iOS app", "4": "Show HN: ClearVoice TTS, SOTA voice cloning model running offline, in an iOS app. This is the first app I am aware of that can run a voice model of this quality on iPhone or iPad. It uses 6GB of RAM at peak, so most modern Macs, and some iPhone and iPad models should work (I\u2019ve tested on a 17 pro). The model used is OmniVoice, which is known mostly for its ability to generate quality tts in hundreds of languages.
This probably isn\u2019t the fastest implementation of OmniVoice on a Mac, but it\u2019s got to be the easiest to run on macOS, and, as far as I know, the only existing implementation on iOS.
Demo of it running on an M3 MacBook Air: https://x.com/RoryClear/status/2090414148030972300", "5": "2026-08-23T07:00:09.886907"} +{"0": 261, "1": "hackernews", "2": "https://apps.microsoft.com/detail/9nkkk5k5s4ct?hl=en-US&gl=US", "3": "Edify \u2013 Windows NLE: OpenFX, proxy editing, AI subs, vtuber tools ($29.99 once)", "4": "Edify \u2013 Windows NLE: OpenFX, proxy editing, AI subs, vtuber tools ($29.99 once). ", "5": "2026-08-23T07:00:16.492806"} +{"0": 265, "1": "hackernews", "2": "https://news.ycombinator.com/item?id=49401293", "3": "Ask HN: If you dislike AI, why don't you prove it?", "4": "Ask HN: If you dislike AI, why don't you prove it?. I've seen so many people who absolutely hate AI, and want AI companies to fail, but then 2 seconds later, open up the coding tool made by frontier labs and willingly hand them money.
Just recently convinced one of them to move away from the frontier lab lock-in to OpenCode and switch between models. Not saying they need to switch to non-frontier models immediately, but this approach lets them easily switch to open source when the time is right.
Why don't more people do this? And shouldn't companies that use AI heavily also suggest this approach?
To be clear, I like AI, and use it all the time, but I use it in a way that allows me to easily switch models and not get locked-in.", "5": "2026-08-23T07:00:16.510603"} +{"0": 279, "1": "hackernews", "2": "https://worthtotry.com", "3": "Show HN: AI tools with real pricing, submitted in 30 seconds", "4": "Show HN: AI tools with real pricing, submitted in 30 seconds. ", "5": "2026-08-23T07:00:16.578270"} +{"0": 280, "1": "hackernews", "2": "https://foundera.app/", "3": "Show HN: AI-Native Accelerator that helps builders make something people want", "4": "Show HN: AI-Native Accelerator that helps builders make something people want. Most people are not validating their ideas before start building it! I am a 5th time tech founder, gaming included and I built & managed startup accelerator programs. I know lot of things about early stage founders and their pains as I have been there. That is why I am building Foundera. It\u2019s an AI-Native Accelerator for tech builders. Today the problem is; most of the AI coding tools opens with a chat and asking \u201ctype your idea and start building\u201d. This is wrong! I want to reduce the time and money spent on ideas that nobody asked for. Try it at https://foundera.app and give me your feedback so that I can help more people.", "5": "2026-08-23T07:00:16.583718"} +{"0": 300, "1": "hackernews", "2": "https://www.cnbc.com/2026/08/03/hugging-face-china-ai-race-open-models.html", "3": "Hugging Face CEO says China is winning the AI race and dominating on open models", "4": "Hugging Face CEO says China is winning the AI race and dominating on open models. ", "5": "2026-08-23T07:00:19.461768"}