Show HN: A package manager for MCP servers https://bit.ly/3RELj5A

Show HN: A package manager for MCP servers https://bit.ly/4jj0Q72 April 9, 2025 at 08:17AM

Show HN: Speech and Audio To Text – Transcribe voice messages offline https://bit.ly/43GPxkA

Show HN: Speech and Audio To Text – Transcribe voice messages offline Hi HN, I’d like to share a side project I recently finished: Speech & Audio To Text, a small iOS utility to transcribe voice messages from apps like WhatsApp, Telegram, and Voice Memos — fully offline and without sending any data to the cloud. The motivation was personal. I often receive voice notes when I can't listen to them — during meetings, while commuting on noisy trains, or when I don’t want to disturb others. I used to rely on an app that handled this well, until it moved to a subscription model and started requiring constant internet access. I wanted something that would: - Work offline - Not send audio to any server - Be available with a single, one-time purchase - Support WhatsApp and Telegram voice messages directly - Handle multiple languages So I built it. It’s completely local, lightweight, and doesn't use any AI or external APIs. The speech recognition relies on on-device iOS capabilities, and supports over 60 languages depending on system settings. I'd love to hear your feedback or suggestions. Thanks for checking it out. https://apple.co/42onqUU April 9, 2025 at 07:15AM

Show HN: My mom was unimpressed with my blog. What do you think? https://bit.ly/4idYiGv

Show HN: My mom was unimpressed with my blog. What do you think? I recently launched my personal tech blog and proudly showed it to my mom. Her response? "It's... fine." Now I need your brutally honest feedback! Blog URL: https://bit.ly/3EeT0MQ Github: https://bit.ly/3R8ItFS PS: Every visitor gets my mom one step closer to understanding what I actually do for a living! https://bit.ly/3EeT0MQ April 9, 2025 at 07:00AM

Show HN: DrawDB – open-source online database diagram editor (a retro) https://bit.ly/3Ra6rRa

Show HN: DrawDB – open-source online database diagram editor (a retro) One year ago I open-sourced my very first 'real' project and shared it here. I was a college student in my senior year and desperately looking for a job. At the time of sharing it i couldn't even afford a domain and naively let someone buy the one i had my eyes on lol. It's been a hell of a year with this blowing up, me moving to another country, and switching 2 jobs. In a year we somehow managed to hit 26k stars, grow a 1000+ person discord community, and support 37 languages. I couldn't be more grateful for the community that helped this grow, but now i don't know what direction to take this project in. All of this was an accident. But now I feel like I'm missing out on not using this success. I have been thinking of monetization options, but not sure if I wanna go down that route. I like the idea of it being free and available for everyone but also can't help but think of everything that could be done if committed full-time or even had a small team. I keep telling myself(and others) i'll do something if i meet a co-founder, but doubt and fear of blowing this up keeps back. How would you proceed? https://bit.ly/4j30fa1 April 9, 2025 at 01:20AM

Show HN: All-in-One Japanese Study Tool with AI + Anki https://bit.ly/42uYBXI

Show HN: All-in-One Japanese Study Tool with AI + Anki I've been using AI a lot lately to help me study Japanese. I can pick up a bunch of words just from watching anime, but I'm still pretty weak when it comes to speaking and reading characters. I tried using ChatGPT to practice, but it quickly got annoying bouncing between different tools — I couldn’t chat and speak at the same time, and juggling stuff like Anki, dictionaries,etc. I built this as an all-in-one tool to do this in one place. Figured there are probably others in the same boat here, so wanted to share it here and see if anyone finds it helpful. Would love any feedback or ideas. Thanks, and hope it helps someone out! https://bit.ly/3G05Wqp April 7, 2025 at 10:10PM

Show HN: Browser-Use MCP for Claude that works without an API key https://bit.ly/44dTCwQ

Show HN: Browser-Use MCP for Claude that works without an API key Hey everyone, I built an MCP that enables the Claude desktop app to control your browser with Browser Use (YC W25). MCPs (based on the Model Context Protocol by Anthropic) enable Claude to use tools that can do a variety of functions like searching the web, executing code, manipulating files, accessing APIs; basically interacting with things outside the conversation context. The main reason I built this MCP was because I saw that existing Browser Use MCPs required me to get an API key from OpenAI/Anthropic playgrounds in order to use them with the Claude desktop app, which I didn't want to do because I already pay for Claude. It was fun to read the Browser Use code ( https://bit.ly/4lgb75H ) to see how everything works! There's basically an agent (powered by an LLM of your choice) which has access to a bunch of tools. I removed that agent and made the Claude desktop app the agent instead using the Model Context Protocol by Anthropic, and finally gave it access to the Browser Use tools to tie everything together. Hope you find it useful! https://twitter.com/vortex_ape/status/1900953901588729864 April 8, 2025 at 12:38AM

Show HN: Turn school textbook into a personal tutor (powered by Groq) https://bit.ly/442DMVZ

Show HN: Turn school textbook into a personal tutor (powered by Groq) https://bit.ly/3FXGnGy April 7, 2025 at 11:48PM

Show HN: A free movies and series randomiser https://bit.ly/4i1Eypi

Show HN: A free movies and series randomiser Hi HN! Over 3 years ago I developed a movies and series randomiser in Cordova web to deploy on app stores. The first versions had ads in it to monetize the development, also a "PRO" version was released for the price of 1 dollar (or something like that, the lowest price tier as far as I remember). Anyway, in recent years I've reached a point where all applications does not need to be monetized, some are just fun to develop and release for free. Show me a Movie is one of them! The current state of the app is the same as of 3 years ago. Besides getting it to work in the browser, minor bug fixes has been fixed. With that said, go ahead and try it out. It is totally free, I only track page count using GoatCounter (for the fun of it). If you want to can leave some valuable feedback and insight which either I or someone else on HN can use in future projects! Warm regards, Lukas https://bit.ly/4cmeCDu April 7, 2025 at 09:00AM

Show HN: Simple time tracking and invoicing for freelancers and small agencies https://bit.ly/4ciWU3S

Show HN: Simple time tracking and invoicing for freelancers and small agencies https://bit.ly/42AK91e April 7, 2025 at 07:34AM

Show HN: I created a ELI5 Blockchain glossary, with also a bit of interactivity https://bit.ly/3G97VZs

Show HN: I created a ELI5 Blockchain glossary, with also a bit of interactivity The reason is simply that I'm learning blockchain and i found myself searching and hardly recalling terminology. So I started building a glossary (that slowly i'm making interactive) in which the terms are explained with ELI5 analogies. AI helped me with that. Hope you find it useful. You can contribute to the data set here: https://bit.ly/4cpFdzK https://bit.ly/41Zoboy April 6, 2025 at 06:10PM

Show HN: I Built ImgFiber-Better Image Optimizer. Free No Limits https://bit.ly/44eWZ6B

Show HN: I Built ImgFiber-Better Image Optimizer. Free No Limits No file size/upload limits. Processed locally right inside your browser. No Server Uploads. NOT A FFmpeg Wrapper. It's not like any other Alternatives, Try for yourself! Are you someone who deals with lots of images and always find yourself with your storage full? or Someone Who deals with websites optimization and would love to get that fast loading speed? No matter who you are, as long as you deal with Images, Imgfiber got your back! ⬇ Reduce image file sizes by up to 95% without losing quality! Supports all major formats: JPG, PNG, JPEG, GIF, SVG, and WebP. Works entirely in your browser—no uploads, no servers, just the power of your own device! Blazing fast compression—processes images as quickly as you drag and drop. Delivers results 2x better than competitors like OptImage, CompressX, TinyPNG, and Squoosh. Totally FREE with no file size or count limits - compress as many images as you want! Social Proof? - I am gonna be honest with you! I don't have social proof! I've been too busy building cool tools like IMGFiber, Codeaway, QuickWrap entire year that i forgot they need Lovely users to have significant value to its existent. I've spent an year building great range of SaaS and had zero exposure to provide your with trusted by 40,000+ users around the world! Nah! I don't have that! that's why i am here! I would geneuinly love for you to give ImgFiber a shot! Check it out ImgFiber.com and let me know your thoughts! https://bit.ly/4coUKzZ April 6, 2025 at 07:55AM

Show HN: Owl, a Spaced Repetition App https://bit.ly/44ape6r

Show HN: Owl, a Spaced Repetition App Owl is built a spaced repetition app. We built it for ourselves mostly because we were unhappy with Anki from a UX perspective, and are now releasing it to everybody else. It is super tiny, but - we think - also pretty good. You can add your own decks manually, or generate them from a PDF (think academic papers, which is how I use that feature) or a prompt. There are no emails except study reminders (when there are cards to study). You can also use our "AI tutor" to review cards conversationally. Looking forward to your feedback! https://bit.ly/42oe1Ni April 3, 2025 at 07:20PM

Show HN: I made a conversational AI for interview prep https://bit.ly/420ZPuF

Show HN: I made a conversational AI for interview prep SpeakFast helps you prep by actually talking—not just reading tips or recording yourself. The AI interviews you, challenges you in real time, and gives coaching based on how you're doing. It can even build a custom roadmap around your weak spots. There are 200K+ real roles from top tech companies to practice with. You can also paste a job description and get a tailored mock interview instantly. It’s composable, flexible, and built to feel real https://bit.ly/41ZrU5v April 5, 2025 at 11:42PM

Show HN: OCR pipeline for ML training (tables, diagrams, math, multilingual) https://bit.ly/42vkSWe

Show HN: OCR pipeline for ML training (tables, diagrams, math, multilingual) Hi HN, I’ve been working on an OCR pipeline specifically optimized for machine learning dataset preparation. It’s designed to process complex academic materials — including math formulas, tables, figures, and multilingual text — and output clean, structured formats like JSON and Markdown. Some features: • Multi-stage OCR combining DocLayout-YOLO, Google Vision, MathPix, and Gemini Pro Vision • Extracts and understands diagrams, tables, LaTeX-style math, and multilingual text (Japanese/Korean/English) • Highly tuned for ML training pipelines, including dataset generation and preprocessing for RAG or fine-tuning tasks Sample outputs and real exam-based examples are included (EJU Biology, UTokyo Math, etc.) Would love to hear any feedback or ideas for improvement. GitHub: https://bit.ly/3EhtxCm https://bit.ly/3EhtxCm April 5, 2025 at 06:22AM

Show HN: Automatic Python shebang lines for venv and conda environment finding https://bit.ly/42mt2ix

Show HN: Automatic Python shebang lines for venv and conda environment finding https://bit.ly/42iH70y April 4, 2025 at 11:51PM

Show HN: Built a visual UI editor that exports to code https://bit.ly/42lKjsg

Show HN: Built a visual UI editor that exports to code I've been specializing in UI for over a decade. Using code for UI is great but I always wished there was a more visual code to build apps and sites. So I went ahead and built it. Decided to build it with Compose Multiplatform since I work a lot with Kotlin. Currently exports to CMP as it was simpler due to the context switch. You can try it for free and export some apps https://bit.ly/3DSaOgQ April 4, 2025 at 11:04AM

Show HN: Transputer emulator in JavaScript (fast enough to be useful) https://bit.ly/42lSDIB

Show HN: Transputer emulator in JavaScript (fast enough to be useful) https://bit.ly/3RxsqBB April 4, 2025 at 04:59AM

Show HN: A new VSCode extension that shows definition functions in a stack https://bit.ly/4lgIRzY

Show HN: A new VSCode extension that shows definition functions in a stack Definition Stack is a new vscode extension I have just released. It is a reading tool for Javascript and Typescript. It is available in the extension marketplace. You just click in a function in any source code, execute a command, and a new tab opens next to the original. That tab has a "block" which contains a copy of the function you clicked in. In that code every word (symbol) that has a definition is highlighted. If you click on a higlighted word a new block opens above the original which contains the source code of the definition for that word. You can click in that block and repeat to create a stack. All function blocks are in the one tab that you can scroll through. There are other options like collapsing a block, deleting it, etc. It is easy to open a block, look at it, and delete it taking you back to the block below. Then clicking in the lower block with another word opens yet another block above. Continuining this process lets you walk the "tree" of references and definitions. This lets you see all the code executed when the original function runs. This is similar to stepping through code with a debugger. The function code is isolated in each block with the name of the original source code file and the lines are numbered the same. But when using the stack you can ignore what file each one came from. This gives a fresh way to look at what all the functions do without the cognitive load of remembering what file the functions are in. AFAIK, this concept is original. Correct me if I'm wrong. My idea came from an IDE for Java from IBM many years ago. In that system there were no source files at all. Each function came from from a database. That IDE was a failure :-) I think my version of that concept will do better because it doesn't replace source files, it just adds a tool for working with source files. The stack is created instantly and is just meant to be used occasionally when it makes sense. It is sort of a Go To Definition on steroids. Please give me feedback. Since it is brand new I want to fix anything wrong including user experience problems. Enjoy ... https://bit.ly/41VK8Vx April 3, 2025 at 11:59PM

Show HN: FlashTokenizer – 10x faster C++ tokenizer for Python https://bit.ly/3Rvb2NM

Show HN: FlashTokenizer – 10x faster C++ tokenizer for Python I built a tokenizer in C++ with a Python binding that outperforms HuggingFace tokenizers by 10x on large inputs. It's optimized for minimal memory usage and latency. Benchmarks and comparison included in README. Would love feedback or contributions! https://bit.ly/4cehmCW April 3, 2025 at 07:46AM

Show HN: OCR pipeline for ML training (tables, diagrams, math, multilingual) https://bit.ly/42dslrT

Show HN: OCR pipeline for ML training (tables, diagrams, math, multilingual) Hi HN, I’ve been working on an OCR pipeline specifically optimized for machine learning dataset preparation. It’s designed to process complex academic materials — including math formulas, tables, figures, and multilingual text — and output clean, structured formats like JSON and Markdown. Some features: • Multi-stage OCR combining DocLayout-YOLO, Google Vision, MathPix, and Gemini Pro Vision • Extracts and understands diagrams, tables, LaTeX-style math, and multilingual text (Japanese/Korean/English) • Highly tuned for ML training pipelines, including dataset generation and preprocessing for RAG or fine-tuning tasks Sample outputs and real exam-based examples are included (EJU Biology, UTokyo Math, etc.) Would love to hear any feedback or ideas for improvement. GitHub: https://bit.ly/3EhtxCm https://bit.ly/3EhtxCm April 3, 2025 at 06:48AM