🛠️ Hacker News Tools

[Other] Show HN: Built an MCP server using Cloudflare's Code Mode pattern Read this article by Cloudflare this morning <a href="https:&#x2F;&#x2F;blog.cloudflare.com&#x2F;code-mode&#x2F;" rel="nofollow">https:&#x2F;&#x2F;blog.cloudflare.com&#x2F;code-mode&#x2F;</a> the main argument being that LLMs are much better at writing typescript code than tool calls because they&#x27;ve seen typescript code many more times.<p>HN Discussion: <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=45399204">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=45399204</a> <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=45386248">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=45386248</a><p>Deno provides a great sandbox environment for Typescript code execution because of its permissions system which made it easy to spin up code that only has access to fetch and network calls.<p>Stick an MCP proxy on top of that and you&#x27;ve got &quot;CodeMode&quot; (code intermixed with MCP tool calls) for more advanced workflow orchestration.<p><a href="https:&#x2F;&#x2F;github.com&#x2F;jx-codes&#x2F;codemode-mcp" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;jx-codes&#x2F;codemode-mcp</a><p>There&#x27;s a lot of things that can be improved here. Like a virtual file system for the agent to actually build up its solution instead of being forced to one shot the solution but the bones are there.

[Other] Show HN: Toolbrew – Free little tools without signups or ads I got tired of fighting through spammy tool sites just to do something simple, so I made Toolbrew in a few hours on Replit. Just a bunch of free little tools in one place. Text converters, SEO checks, video downloaders, that kind of stuff. No signups, no ads.<p>If there is a tool you wish existed, you can request it on the site and I will build it. Do your worst. Seriously, ANY tool.<p>Maybe it helps, maybe not. Enjoy!

[Other] Show HN: Curated gamedev specific search engine

[Other] Show HN: An open source Launchpad for macOS 26 An open sourced version of Launchpad that I enhanced, you can import from old system Launchpad (just one click), and most of things look same. There are also a lot of functions, like adjust &#x2F; display &#x2F; hide icon size and title, localize icons, game controller support, i18n, voice over, etc.<p>Still updating, open an issue if there are any problems. Hope this could help if someone updated to MacOS26 and not happy with new Launchpad :)<p>Thank you.

[API/SDK] Show HN: One API for all your SMTPs Hi HN,<p>I host my own mail infrastructure and don&#x27;t rely on third-party tools to send my mails. But in a microservice architecture, managing all the SMTP configs quickly becomes tiresome. After a while I&#x27;d always have my credentials and settings scattered in application-properties, bash scripts and environment files.<p>So I&#x27;ve build Brieferl, an app where you can add your SMTP servers and send emails through a single API with a simple JSON payload. You also get logs of when&#x2F;where emails were sent and HTML previews of messages.<p>I am interested in what you guys think about this. You can create a free account with just your email, add a SMTP server + API key and start sending. (There is no upsell, or paid plan yet)<p>This is super early, it works and I made it actually just for myself but a friend told me he‘d also love to use this, so I thought why not ask other developers what they think.<p>Is this something you’d use? Are there some features that would it more valuable for you? Or is this something only I’ve felt?

[Other] Property-Based Testing of OCaml 5's Runtime System [pdf]

[Other] Show HN: A web version of Pips game (NYT domino game) Hi everyone,<p>I’m an indie developer learning Next.js and a big fan of the NYT game Pips. Inspired by it, I built <a href="https:&#x2F;&#x2F;pipsgamer.com" rel="nofollow">https:&#x2F;&#x2F;pipsgamer.com</a> — a responsive web version of Pips with smooth gameplay on both desktop and mobile.<p>What makes this project different from NYT’s version is that you can play it infinitely under three difficulty levels: Easy &#x2F; Medium &#x2F; Hard.<p>This is the first time I’ve built a game. Along the way I ran into many difficulties: implementing the game logic, configuring the UI, matching layouts for small and large screens, etc. I spent many lonely nights and sometimes even doubted whether I could complete the whole project. After 24 days of persistent effort, the project is finally finished.<p>No signup required — just go and play. If you try it out I’d really appreciate your feedback: what you like, what bugs you see, what could be improved.<p>Thanks!

[Other] Show HN: FingerprinterJS – A tiny JavaScript library for browser fingerprints I made FingerprinterJS, a small library with no dependencies that creates browser fingerprints from signals like canvas, WebGL, audio, fonts, userAgent, and screen info.<p>It’s written in TypeScript, lets you enable&#x2F;disable collectors, add custom data, and includes a simple suspicious-activity score.<p>Would love feedback.

[Other] Genode OS Framework

[Other] Show HN: Macscope – I decide to built a better Cmd-Tab replacement for macOS Hi HN,<p>Macscope is a new window manager and and app switcher for macOS built on the philosophy of enhancing, not replacing, your existing muscle memory.<p>It works by augmenting the familiar Cmd+Tab workflow. A quick tap of your shortcut instantly switches between recent apps, just like you&#x27;re used to. A slightly longer hold, however, opens the full Macscope interface where you can manage all your open windows and tabs.<p>You can also use modifier shortcuts to enter Placement Modes, which let you instantly snap a selected window to the left&#x2F;right&#x2F;top&#x2F;bottom&#x2F; half of your screen.<p>Here are some of the key features:<p>- Unified Search &amp; Switch: A single interface to instantly find and switch to any window, browser tab (Safari, Chrome, Arc, etc.), or application just by typing.<p>- Live Previews: See a real-time preview of what&#x27;s inside each window so you know exactly where you&#x27;re going. You can also disable previews for a more minimal experience.<p>- Advanced Window Management: Go beyond just switching. Select multiple windows and arrange them into layouts like vertical&#x2F;horizontal splits or grids.<p>- Scopes: Save collections of app windows as a &quot;Scope&quot; and instantly restore that entire workspace later. It&#x27;s ideal for quickly switching between different projects or tasks.<p>It’s a native macOS app built with Swift and supports both Apple Silicon and Intel machines.<p>Launch Offer for HN:<p>There&#x27;s a free trial with 250 actions. For the Hacker News community, I&#x27;m offering a 50% discount on the lifetime license.<p>Website: <a href="https:&#x2F;&#x2F;macscope.app" rel="nofollow">https:&#x2F;&#x2F;macscope.app</a><p>Discord Community: <a href="https:&#x2F;&#x2F;discord.gg&#x2F;ehktEWr97K" rel="nofollow">https:&#x2F;&#x2F;discord.gg&#x2F;ehktEWr97K</a><p>I&#x27;ll be here all day to answer questions and would be grateful for any feedback. Thanks for checking it out!

[API/SDK] Show HN: Prism – Let browser agents access any app Hey HN, We’re Alex, Land, and Rajit. We’re building Prism (prismai.sh), a tool that helps browser agents authenticate onto websites with user credentials. Developers pass in credentials, Prism logs into a website on their behalf, and hands them back the cookies so they have an authenticated session. Here’s an example of how developers can use Prism to complete username&#x2F;password flows (<a href="https:&#x2F;&#x2F;youtu.be&#x2F;SEtVUnWnxuE" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;SEtVUnWnxuE</a>), and here’s an example of how developers can use Prism to complete login flows that require an OTP code (<a href="https:&#x2F;&#x2F;youtu.be&#x2F;fe9w9PvrwH0" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;fe9w9PvrwH0</a>).<p>We spoke to browser agent developers and saw people copying and pasting credentials and even credit card numbers directly into model system prompts. We were surprised that there wasn’t a better way to give agents access to websites on a human’s behalf. Moreover, we noticed that every company had to build infrastructure to manage OTP, TOTP, and MFA and that auth remained a significant hurdle in agent reliability. We wondered if this was a boring part of the problem of building web automations that someone could automate away.<p>We started working with Casco, an autonomous security testing company, to enable their agent to access customer sites. Before a pentest, Casco makes a request to Prism’s API specifying test user credentials, a domain, and a login method. For example, give me an authenticated session for the account rajit@prismai.sh for OpenAI via OTP code over email. Our agent logs in on their behalf (without exposing credentials to a model), and we download the cookies and send them back in the response.<p>To maintain speed and reliability, we use playwright in most cases to login (which gives us speed), and we fallback to AI on failure (which gives us reliability). We have a number of websites we support out of the box and add new scripts as the number of websites we need to support grows. We are working on a way for the agent to update the existing playwright script on failure, so our scripts always stay up to date.<p>To try our api, you can use our API playground docs.prismai.sh&#x2F;api-reference&#x2F;endpoint&#x2F;login to sign into x.com with the following API key: pk_54abb1cd0a637eb973ed690416e71a953e98f2ea839cf16529bbfa41a41bc016 .<p>We’d love to learn more about how other developers give agents access to their accounts. We look forward to everyone’s feedback and comments.

[Other] Show HN: Phishcan, Canada's first open and free threat intelligence platform Phishcan provides crucial threat intelligence, and it currently tracks phishing domains for:<p>• Scotiabank, Desjardins, RBC, Interac…<p>• Telecom providers, provincial power and health services...<p>• Federal &amp; provincial services, CRA, Canada Post, Service Canada, Revenue Québec...<p>How Phishcan works:<p>• Parsing millions of domains: Continuously scanning and analyzing vast numbers of domains to detect suspicious patterns and potential phishing sites.<p>• Monitoring threat actors : close watch on cyber‑criminal infrastructures and their new domain registrations.<p>• Data enrichment : adding contextual insights and connections to improve the information<p>• Feeds are updated every 12 hours.<p>• You can use the API freely at: <a href="https:&#x2F;&#x2F;phishcan.com&#x2F;api-docs" rel="nofollow">https:&#x2F;&#x2F;phishcan.com&#x2F;api-docs</a><p>Data is also available on: <a href="https:&#x2F;&#x2F;github.com&#x2F;Phishcan&#x2F;phishcan-data" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;Phishcan&#x2F;phishcan-data</a><p>I plan to improve the whole platform with more data during my free time!

[IDE/Editor] Rustroid, a Rust IDE for Android

[CLI Tool] Better Curl Saul: a lightweight API testing CLI focused on UX and simplicity

[Other] Show HN: A UI Library for the Web Focusing on accessibility, longevity, performance, and simplicity

[Other] Show HN: Dayflow – A git log for your day Hi HN! I&#x27;ve been building Dayflow, a macOS app that automatically tracks what you&#x27;re actually working on (not just which apps you have open).<p>Here&#x27;s what it does:<p>- It creates a semantic timeline of your day;<p>- It does it by understanding the content on your screen (with local or cloud VLMs);<p>- This allows you to see exactly where your time went without any manual logging.<p>Traditional time trackers tell you &quot;3 hours in Chrome&quot; which is not very helpful. Dayflow actually understands if you&#x27;re reading documentation, debugging code, or scrolling HN. Instead of &quot;Chrome: 3 hours&quot;, you get &quot;Reviewed PR comments: 45min&quot;, &quot;Read HN thread about Rust: 20min&quot;, &quot;Debugged auth flow: 1.5hr&quot;.<p>I was an early Rewind user but rarely used the retrieval feature. I built Dayflow because I saw other interesting uses for screen data. I find that it helps me stay on track while working - I check it every few hours and make sure I’m spending my time the way I intended - if I’m not, I try to course correct.<p>Here’s what you need to know about privacy:<p>- Run 100% locally using qwen2.5-vl-3b (~4GB model)<p>- No cloud uploads, no account<p>- Full source available under MIT license (<a href="https:&#x2F;&#x2F;github.com&#x2F;JerryZLiu&#x2F;Dayflow" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;JerryZLiu&#x2F;Dayflow</a>)<p>- Optional: BYO Gemini API key for better quality (stored in Keychain, with free-tier workaround to prevent training on your data)<p>The tech stack is pretty simple, SwiftUI with a local sqlite DB. Uses native macOS apis for efficient screen captures. Since most people who run LLMs locally already have their tool of choice (Ollama, LLMStudio, etc.), I decided to not embed an LLM into Dayflow.<p>By far the biggest challenge was adapting from SOTA vision models like Gemini 2.5 Pro to small, local models. My constraints were that it had to take up &lt;4GB of ram and have vision capabilities. I had to do a lot of evals to figure out that Qwen2.5VL-3B was the best balance of size and quality, but there was still a sizable tradeoff in quality that I had to accept. I also got creative with sampling rates and prompt chunking to deal with the 100x smaller context window. Processing a 15 minute segment takes ~32 local LLM calls vs 2 Gemini calls!<p>Here’s what I’m working on next:<p>Distillation: Using Gemini&#x27;s high-quality outputs as training data to teach a local model the patterns it needs, hopefully closing the quality gap.<p>Custom dashboards where you can track answers to any question like &quot;How long did I spend on HN?&quot; or &quot;Hours until my first deep work session of the day<p>I&#x27;d love to hear your thoughts, especially if you&#x27;ve struggled with productivity tracking or have ideas for what you&#x27;d want from a tool like this.

[DevOps] Show HN: Mosaic – A Kotlin framework for cleaner back end code Backend APIs often grow into large orchestration classes full of duplicated calls and manual concurrency.<p>I’ve been working on Mosaic, a Kotlin framework that composes responses out of small, request-scoped “tiles.” Each tile runs once per request, dependencies resolve automatically, and independent tiles execute in parallel without boilerplate.<p>It’s still early (v0.2.0), but working today for caching, concurrency, and testability. Curious to hear feedback on the approach.<p>GitHub: <a href="https:&#x2F;&#x2F;github.com&#x2F;Nick-Abbott&#x2F;Mosaic" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;Nick-Abbott&#x2F;Mosaic</a> Maven Central: org.buildmosaic:mosaic-core:0.2.0

[Other] Show HN: Verdent – AI coding agent that plans, tests, and ships Hi HN!<p>We just launched Verdent, an AI coding agent that actually plans before it codes, tests its own work, and delivers production-ready results without you having to clean up after it.<p>tl;dr Check out the website [1] and read the story [2]<p>- The problem: Current AI tools dump code and leave you to debug. You ask for &quot;build a login endpoint&quot; and get a passwordless JWT mess that doesn&#x27;t fit your project. We&#x27;ve talked to hundreds of engineers who are frustrated with having to micromanage AI output.<p>- What we built: Verdent works more like a senior engineer. It asks clarifying questions first, breaks down complex tasks, writes the code, then runs its own tests and fixes bugs until everything works. We call it the Plan → Code → Verify loop.<p>Two ways to use it: VS Code extension for staying close to the code, or Verdent Deck desktop app for handling bigger multi-component tasks in parallel.<p>- Background: I was Head of Algorithms at TikTok and spent 9 years as Chief Technical Architect at Baidu before this. Building those massive recommendation systems taught us that you need hundreds of specialized models working together, not just wrapping an LLM. We applied the same system engineering approach to AI coding.<p>- Try it: Available now, paid plans start at $19&#x2F;month. We&#x27;re a small team so no free tier, but we&#x27;d rather you experience what autonomous coding can actually do than get frustrated with a limited version.<p>- The bigger picture: We see AI coding as just the start of a larger shift where AI becomes the orchestrator across the entire software lifecycle, not just a typing assistant.<p>What do you think?<p>What tasks would you actually trust an AI agent to handle end-to-end?<p>[1] <a href="https:&#x2F;&#x2F;www.verdent.ai&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.verdent.ai&#x2F;</a> [2] <a href="https:&#x2F;&#x2F;thenewstack.io&#x2F;tiktoks-ex-algorithm-chief-launches-verdent-ai-coding-tool&#x2F;" rel="nofollow">https:&#x2F;&#x2F;thenewstack.io&#x2F;tiktoks-ex-algorithm-chief-launches-v...</a>

[Other] Amiga SPICE is a program for simulating electronic circuits

[Other] Python developers are embracing type hints