标签：#p

GitHub热门项目 | You're the boss, agents are your team. They handle tasks on their own, message each other, and review each other's work. You just watch the kanban board and give high-level commands. Codex/Claude/OpenCode/Cursor/Grok/GitHub Copilot/Kiro/Z.AI/MiniMax/Kimi(200+ models, 75+ LLM providers, free models no auth). Build your AI company with multiple teams | Stars: 1,680 | 23 stars today | 语言: TypeScript

GitHub热门项目 | A progressive Node.js framework for building efficient, scalable, and enterprise-grade server-side applications with TypeScript/JavaScript 🚀 | Stars: 76,210 | 15 stars today | 语言: TypeScript

GitHub热门项目 | Free AI with your ChatGPT account | Stars: 933 | 109 stars today | 语言: TypeScript

GitHub热门项目 | MCP servers for the UniFi suite of applications, Network, Protect, Access, and Drive | Stars: 579 | 14 stars today | 语言: Python

GitHub热门项目 | 🔥🔥🔥 AI-driven database tool and SQL client, The hottest GUI client, supporting MySQL, Oracle, PostgreSQL, DB2, SQL Server, DB2, SQLite, H2, ClickHouse, and more. | Stars: 26,906 | 399 stars today | 语言: Java

This is a submission for DEV's Summer Bug Smash: Smash Stories powered by Sentry . Written with the help of AI (Claude). The bug, the fix, the validation setup, and every claim below are mine, and were verified against the real codebase and a real full disk. The report Someone lost a Magic: The Gathering decklist. They were playing on Cockatrice — the open-source MTG client — with their decks on a drive that had quietly filled up while Oracle pushed an update in the background. They added a card, hit Ctrl+S, and Cockatrice said it saved. The debug log agreed: [2026-05-28 22:31:42.031 I] Saved deck to "G:/cockatrice300/data/decks/edh-b2-gitrog-reanimate.cod" with format 1 - true - true . Success. The file was 0 bytes. The deck was gone. That was issue #6952 , filed by Mekkiss. The steps to reproduce are four lines long and completely damning: Have a full disk. Open a deck on the full disk Add one card to it Save the deck (ctrl+s) Observe that the deck is now a 0 byte file. Three ways to be wrong at once The save path lived in DeckLoader::saveToFile() . Stripped down, it looked like this: QFile file ( fileName ); if ( ! file . open ( QIODevice :: WriteOnly | QIODevice :: Text )) { qCWarning ( DeckLoaderLog ) << "Could not create or open file:" << fileName ; return std :: nullopt ; } bool success = false ; switch ( fmt ) { /* ... saveToFile_Native / saveToFile_Plain ... */ } file . flush (); file . close (); qCInfo ( DeckLoaderLog ) << "Saved deck to " << fileName << "with format" << fmt << "-" << success ; There are three independent failures stacked on top of each other here, and you need all three to lose data: 1. WriteOnly truncates on open. The instant open() succeeds, the existing deck is 0 bytes. Not after a successful write — at open time . The old deck is already destroyed before a single byte of the new one is written. On a full disk, open() still succeeds: truncating a file doesn't need free space. It frees space. 2. The serializers always returned true . sa

Have you ever noticed that your GenAI applications are spending massive amounts of time and money re-reading the exact same setup text? Every time a user asks a short question in a chatbot, the Large Language Model (LLM) must re-read your entire 2,000-word corporate playbook, your agent's system rules, and the full chat history from scratch. This phase is called the pre-fill math phase, and it drives up both your cloud bill and your user latency (Time-to-First-Token).With Amazon Bedrock Prompt Caching for Claude 4.6 (both Sonnet 4.6 and Opus 4.6), this problem is completely solved. You can achieve up to a 90% cost reduction on input tokens and an 85% drop in latency by using a clever architectural shortcut. Here is exactly how it works under the hood, how AWS maintains it across API requests, and how to implement it using Python. The Secret Architecture: Model Inference vs. AWS Infrastructure Prompt caching is a beautiful team effort between the AI model hardware and the AWS cloud infrastructure. The Model Level (The Brains): Inside Claude 4.6, text is processed through mathematical matrices called KV (Key-Value) Caches. Instead of re-reading text, the GPUs calculate the meaning of your system instructions once and build a "mathematical profile." When a cache point is triggered, the model freezes this calculated KV state inside the GPU memory. 2.The AWS Bedrock Level (The Manager): Normally, an LLM wipes its memory the millisecond an API call finishes. AWS Bedrock changes this. It takes your static prompt, creates a secure, unique cryptographic hash (fingerprint), and pins that KV memory block alive. When your next API request comes in, AWS Bedrock instantly hashes the new incoming prompt text. If the top section matches a saved fingerprint, Bedrock's router bypasses the standard pre-fill setup and routes your request directly to the GPU holding your frozen mathematical profile. It is exactly like loading a "Save Game" file instead of restarting a video game from Le

Hey folks 👋 If you've ever wanted to save a video lecture, a livestream replay, or a podcast episode for offline listening, you've probably run into the usual options: sketchy "online video parser" websites that ask you to paste your link into their server, or desktop apps that want you to sign up and upload stuff. Neither feels great when the whole point is your content. I went looking for something better and ended up working with FlowPick — an open-source, privacy-first media downloader that runs entirely in your browser. No uploads, no accounts, no telemetry. Everything (sniffing, downloading, merging, transcoding) happens client-side with FFmpeg compiled to WebAssembly. In this tutorial we'll: Clone and run FlowPick locally Download our first HLS ( .m3u8 ) and DASH ( .mpd ) stream Build and deploy it Poke at the internals so we can customize it If you just want to try it without installing anything, there's a hosted version at https://flowpick.net (more below). The full source is on GitHub: https://github.com/ezwebtools/flowpick . 🔗 Repo: https://github.com/ezwebtools/flowpick · Live tools: https://flowpick.net A 30-second primer: what are HLS and DASH? Before we touch code, two words you'll see everywhere in this space: HLS (HTTP Live Streaming) uses a .m3u8 manifest that lists small .ts (or fMP4) segments. Common for live streams and a lot of video platforms. DASH (Dynamic Adaptive Streaming over HTTP) uses a .mpd manifest; video and audio usually travel as separate .m4s tracks. YouTube and Bilibili lean on this. The key idea: the "video" isn't one file. It's a playlist pointing at dozens (sometimes hundreds) of tiny segments. A downloader's job is to fetch all the segments, decrypt them if needed, and stitch them back into one playable file. That's exactly what FlowPick does — in the browser. What FlowPick is, in one paragraph FlowPick is a Nuxt 4 app that ships in two shapes: A browser extension that sniffs media from the current tab's network requests. An

Nobody has ever asked you to upstream your dotfiles. For thirty years that was the deal with every tool we touched: if you didn't like it, you changed it, and the change lived with you. Then the tools started writing the code, and the deal quietly ended. This essay is about why the deal is back on the table. Because the thing that used to make forking expensive — the labour — is exactly what agents just made cheap. In This Article The Workbench Instinct Then the Harness Era Arrived Forking Was Always the Escape Hatch Code Got Cheap What I Shipped Into My Fork A Feature Does Not Have to Be Useful Your Desire Is the Limit The Fork Is the Destination Now the Discipline Part Bring Back the Joy The Workbench Instinct Show me a developer who has never touched their config and I'll show you someone who hasn't started yet. Vim users brag about their init.lua the way woodworkers talk about a hand plane they've had for twenty years. Emacs people wrote a whole operating system inside a text editor because they could. VS Code won partly because it shipped an extension API and got out of the way. Dotfiles repos are public artifacts, starred and forked, because the setup is part of the craft. This isn't productivity theatre. Some of it is genuine need, some of it is fixing a specific annoyance that only you have, and a lot of it is just fun. All three are valid. The workbench is where the joy lives — and nobody ever waited for permission to alias a command. Then the Harness Era Arrived Then agentic coding tools showed up and quietly changed the shape of the deal. The best-in-class agent harnesses are increasingly vendor-controlled. Claude Code is a product, not a repo you can clone and rebuild. Google announced it's retiring Gemini CLI in favour of a closed-source successor. And note where the line falls: Codex CLI is Apache-2.0 and sitting right there on GitHub, but the Codex desktop app — the thing most people actually click on — is not. The terminal stayed open. The interface

Why browser-based? Every other "online tool" site uploads your PDFs and images to their server for processing. That means: Your sensitive documents sit on someone else's machine Processing speed depends on their server load File size limits, watermarks, or forced signups I built everything using client-side processing — Canvas API, pdf-lib, Tesseract.js (WebAssembly), and more. Your files literally never leave your device. What's included PDF Tools (28): Merge, split, compress, convert to Word/Excel, password protect, watermark, page numbers, metadata editor Image Tools (30): Compressor, background remover, crop, resize, DPI changer, EXIF viewer/remover, OCR (image to text), meme generator, QR code generator Developer Tools (42): JSON formatter, JWT decoder, Base64/Base32 encoder, regex tester, cron generator, SQL formatter, CSS/JS/HTML minifier Design Tools (14): CSS gradient, box shadow, border radius generators, color contrast checker (WCAG), Tailwind component builders Calculators (16): EMI, SIP, BMI, compound interest, salary tax, GST/VAT, ROI, fuel cost Plus text tools, unit converters, YouTube tools, utilities, and more. Tech Stack Next.js 15 (static export, deployed on Cloudflare Pages) TypeScript (strict mode) Tailwind CSS (dark mode) pdf-lib, pdfjs-dist (PDF processing) Tesseract.js (OCR — WebAssembly, zero dependency on external APIs) @imgly /background-removal (on-device AI background removal) SEO and Performance Every tool page includes: FAQ + HowTo structured data (visible in Google rich results) BreadcrumbList schema BlogPosting schema for blog articles Keyword-optimized titles and descriptions OG images for social sharing next/image with priority hints for fast LCP Try it toolshubs.app Looking for feedback — especially on the image compressor, PDF merger, and background remover. What tools would you add next?

How synthetic reality may influence human intuition about motion, gravity, and causality AI video generation has reached a point where a model can create scenes that look physically convincing at first glance: A person jumping impossible distances Objects moving without inertia Water flowing upward Animals performing human-like actions Buildings bending like rubber People interacting with impossible environments For decades, humans learned physics by observing the real world. A ball falls. A glass breaks. A person cannot walk through a wall. Heavy objects require more force to move. These observations create what cognitive scientists call intuitive physics : an internal mental model that predicts how objects should behave. But what happens when the majority of visual experiences become synthetic? Could AI-generated videos slowly change how future generations perceive reality? Humans Do Not See Reality Directly A common misconception is that our brain works like a camera: Reality → Eyes → Brain → Understanding The actual process is closer to: Reality ↓ Sensory input ↓ Brain prediction model ↓ Perception The brain is constantly predicting what should happen next. When you see a ball thrown into the air, your brain automatically predicts: trajectory speed gravity collision point acceleration This happens before conscious reasoning. This capability is known as predictive processing . Your brain is not only asking: "What am I seeing?" It is also asking: "Does this match my internal model of how the world works?" The Brain Learns Physics From Experience Young children do not learn physics from equations. They learn by interaction. A baby discovers: Objects continue to exist when hidden Unsupported objects fall Solid objects cannot overlap Larger objects require more effort to move Researchers call these abilities core knowledge systems . Humans appear to have an innate expectation that the physical world follows consistent rules. For example: A child watching a ball roll

The problem Every time I needed to hand someone a quick installable build — a client, a tester, myself on a different machine — I had to either keep a full local toolchain ready (Android Studio, Flutter SDK, Visual Studio...) or spend 20 minutes reinstalling one just for a single build. So I built PhilBuilder : upload a zipped source project, pick a platform, get back an installable app. No local setup required. 🔗 Try it: https://philbuilder.netlify.app What it does You upload a .zip of your project. The tool: Auto-detects the project type (React, Vue, Flutter, React Native, Kotlin, .NET MAUI, Python, Go, Godot, and 14 others — 22 combinations total) Builds it on remote CI Gives you a download link for an APK, AAB, or Windows .exe No account needed for occasional use (3 builds/day). A free account bumps that to 10/day. How it's built The stack is intentionally simple: Frontend : a single static HTML file, no framework, no build step Backend : a Cloudflare Worker handling auth, rate limiting, and dispatching builds Build execution : GitHub Actions — one big workflow with per-language jobs (Cordova for web frameworks, Capacitor for modern web, native Gradle for Kotlin/Java, dotnet publish for MAUI, flutter build for Flutter, briefcase for Python, gomobile for Go, etc.) Storage : Cloudflare R2 for source zips and build artifacts The auto-detection logic walks the extracted zip looking for telltale files — pubspec.yaml → Flutter, *.csproj → .NET MAUI, capacitor.config.* or a @capacitor/core dependency → Capacitor, build.gradle without package.json → native Kotlin/Java, and so on — with fallbacks down to plain HTML. Some technical details Signing : for Android release builds, it can auto-generate a keystore (and let you download it afterward — losing it means you can never update your app on Play Store again, so this is clearly flagged) or accept an uploaded one. Windows builds : this is the newest addition. Flutter and .NET MAUI need windows-latest runners; Go cross-com

Our top-pick sleeping pads from Nemo, Therm-a-Rest, and Gossamer Gear use high-tech materials to engineer the best sleep you can get outdoors.

Tired of eye strain? These tips will help optimize your smartphone for marathon reading sessions.

Motorola phones may seem old-school, but their reasonable prices, colorful designs, and simple software make them good, wallet-friendly Android smartphones.

You’re an action hero, and you need a camera to match. We guide you through all the models, plus accessory recommendations and hidden software tricks to try.

You don’t have to upload your video to the cloud or pay a monthly fee to secure your home. These security cameras record locally.

Opus 5 runs at 5 and 25 per million tokens against Fable 5 at 10 and 50, so the top tier now costs double for a much smaller gap Thinking is on by default on Opus 5, which silently changes what a tight max_tokens setting does to your output Disabling thinking now returns an error above high effort, so any xhigh or max route that turns it off needs an audit before you migrate Prompt caching starts at 512 tokens on Opus 5, half the Opus 4.8 floor, so short reusable prompts cache with no code change In June I worked through whether Claude Fable 5 was worth double the price of Opus 4.8 and concluded that it usually was, for hard work. Claude Opus 5 landed on July 24 at Opus 4.8's exact price and closed most of that gap. So the answer changed, and a few of the changes will throw errors in code that worked last week. The Price Gap Held, the Capability Gap Closed Opus 5 costs 5 and 25 per million tokens, input and output. That is identical to Opus 4.8 and exactly half of Fable 5 at 10 and 50. Anthropic did not raise the sticker price on the tier it improved, which is the single most consequential fact in this release. What that buys, on the numbers: 79.2 percent on SWE-bench Pro against Fable 5's 80.3, and a CursorBench 3.2 result Anthropic describes as landing within 0.5 percent of Fable 5's peak at max effort, at half the cost per task. On OSWorld 2.0 it goes past Fable 5's best computer-use result at just over a third of the cost. A 1.1 point deficit on the headline coding row, for half the money. Last month the equivalent comparison had an 11 point spread. That is what actually changed, and it flips the default: Fable 5 used to be the reasonable choice for anything hard, and now it has to argue for itself on each task. There is a quieter cost lever too. The minimum cacheable prompt on Opus 5 is 512 tokens, down from 1024 on Opus 4.8. Prompts I had written off as too short to cache now create entries with no code change at all. If you run a lot of small repeated calls,

Three labs shipped flagship models in fifteen days: GPT-5.6 Sol on July 9, Kimi K3 on July 16, Claude Opus 5 on July 24 Opus 5 leads SWE-bench Pro 79.2 to 64.6 over Sol, and ARC-AGI-3 30.2 to 7.8 Sol holds Terminal-Bench 2.1 at 91.9 percent in its top mode and still takes DeepSWE 1.1 and HealthBench Professional Kimi K3 is a 2.8 trillion parameter open-weight model at 3 and 15 per million tokens, roughly 40 percent under Opus 5 on input Fifteen days. That is the gap between OpenAI making GPT-5.6 Sol generally available and Anthropic shipping Claude Opus 5, with Moonshot dropping a 2.8 trillion parameter open-weight model in the middle of it. I wrote a frontier check like this in June and most of it is already out of date, so here is where the three current flagships actually stand. Three Flagships in Fifteen Days Model Lab GA Context Per million (in / out) GPT-5.6 Sol OpenAI 2026-07-09 1.05M 5 / 30 Kimi K3 Moonshot AI 2026-07-16 1M 3 / 15 Claude Opus 5 Anthropic 2026-07-24 1M 5 / 25 The specs have converged to the point where they barely differentiate anything. All three sit at or just above a million tokens of context. All three cap output around 128k. The input prices are within a factor of two of each other. Two years ago a context window was a headline; now it is table stakes, and the interesting differences have moved entirely into behavior under load. Two timing details that get flattened in the coverage. GPT-5.6 Sol was previewed on June 26 and only became generally available on July 9, so some of the earliest benchmark tables were run against a preview build. And Sol is the top of a three-model family alongside Terra and Luna, spanning roughly 1 to 30 per million tokens depending on tier. Comparing Opus 5 to "GPT-5.6" without saying which one is close to meaningless, which is a large share of the comparisons currently circulating. One structural note on Kimi K3, because the parameter count gets quoted carelessly. It is a mixture-of-experts model with 896 exp

Every few weeks I'd end up rewriting the same 10 things from scratch: rate limiter middleware, webhook signature check, retry-with-backoff, connection pool config. So I built AutoSnippets. 50 snippets across Python, JS, TS, Java, C#, C++, Go, PHP, Rust, and SQL. All production-ready, and even more are being made. Favorites of mine: Go channel-based worker pool (snippet #32) Rust Arc + Mutex safe counter (#41) SQL recursive CTE for org charts (#50) PHP RBAC in like 8 lines (#38) Free, no signup. Bookmark it if you find it useful.