🛠️ All DevTools

[Other] Flutter makes it easy and fast to build beautiful apps for mobile and beyond

[Other] Minimus container images are now free

[Other] Haystack: Open-Source AI Framework for Production Ready Agents, RAG

[Other] Show HN: WebBase-III – dBASE III rebuilt in the browser with its own interpreter

[Other] Show HN: Y – A malleable coding-agent desktop app built with Electron

[Other] FEXPRs vs. vtable: how LispE interpreter works

Anthropic updates their terms to verify age or identity

[Testing] Show HN: RLM-based local debugger for AI agent traces We built HALO (Hierarchal Agent Loop Optimizer), an open-source tool for debugging and optimizing AI agents using their execution traces.<p>It’s a loop. Run your agent, feed the traces to HALO, get the report, apply the fixes, then re-run your agent.<p>HALO takes in OTEL compliant traces from AI agents using tracing frameworks such as Langfuse, Arize&#x2F;OpenInference, or even just plain JSONL. It uses an RLM (Recursive Language Model) to more efficiently break trace analysis into smaller subproblems in order to find recurring patterns across large amounts of data and fix systemic issues that regular LLMs might typically miss.<p>You can also optionally provide a path to where your agent code lives to give the engine more context so it can more concretely provide useful insights.<p>The repo also includes a desktop app that you can run locally without having to sign up for anything or configure anything complex.<p>Check out the readme in the repo for more in depth information on what HALO is and how you can use it to your benefit :)

Fired by Google for creating the Google workspace CLI <a href="https:&#x2F;&#x2F;xcancel.com&#x2F;JPoehnelt&#x2F;status&#x2F;2069482265953087602" rel="nofollow">https:&#x2F;&#x2F;xcancel.com&#x2F;JPoehnelt&#x2F;status&#x2F;2069482265953087602</a>

[Other] Building apps using TanStack Start

San Diego photologs from the 1970s

[Other] Show HN: The Cascade Graph – An interactive map of AI and energy constraints Hello, I wanted to share with you all a interactive map of the economics and physics constraints of the AI buildout. It has macro drivers, industrial chokepoints, and where that shows up in markets.<p>I&#x27;ve added 393 nodes and 562 edges to capture other supply &#x2F; physics constraints as well.<p>There&#x27;s no sign up, and no pay wall, it&#x27;s all free.<p>Please let me know what you think!

[Other] Show HN: peerd – AI agent harness that runs entirely in your browser Hey HN. <a href="http:&#x2F;&#x2F;peerd.ai" rel="nofollow">http:&#x2F;&#x2F;peerd.ai</a> is an AI agent harness that lives entirely in your browser as a web extension. You don’t have to install a separate “AI browser”. You don’t have to bolt on or run some external process or manage a clunky mcp integration. It’s just a fully contained web extension, written in no build vanilla JS with minimal non-browser dependencies, using your own provider keys, and Apache 2.<p>This isn’t just a fun hack. While it has largely been a solo side project, I genuinely believe the browser and the web could be the most natural platform for AI agents to operate safely, autonomously, and most importantly without A2A middlemen (more on that in a sec). To demonstrate that point peerd doesn’t just drive browser automation. It spins up isolated sandboxes using tabs and worker instances to support various real workload types. Those include headless JS computational work, visual JS notebooks, personal client side apps, and real Linux VMs on top of wasm with full http networking.<p>The industry discourse over the last several months has been dominated by “which substrate is the best for ai agent sandboxes” with many competing answers focused on different models and use cases. Cloudflare is one of the most prominent examples, positioning its v8 isolate based workers as the best in class solution thanks to faster than container startup times and strong isolation guarantees. The v8 isolate is of course the product of chromium, which runs on billions of browsers around the world for free. The browser as a whole is perhaps the most battle tested sandbox system in the entire software industry. It’s been built on 3 decades of learning from hostile content, hostile code, and hostile users. Native and cloud agents are necessarily rebuilding all or most of this posture from scratch. peerd doesn’t. It leverages everything the browser has to offer and pushes it to its functional limits, while inheriting its security baseline and isolation from the host system.<p>Robust sandboxing isn’t the only thing the browser offers and peerd uses. It comes with extremely powerful and underrated primitives, from webCrypto, webRTC, webAuthn, webGPU, and ~soon WebNN. Direct web access, with your real live sessions, and api calls with fetch present an alternative model to MCP integrations. The agent can write and spawn web apps right there in a tab, no hosted service necessary. Then there’s the A2A piece: peerd already has a rudimentary p2p (peerd-to-peerd?) network in place using webRTC. Today you can connect with peers on the network, add them as contacts, and share signed apps you’ve created. I’m working on extending these apps to be able to leverage the same p2p network to support decentralized web apps (dwapps), as well as facilitate true p2p A2A with no platform or middlemen.<p>Given this is an early part time project, this is an extremely experimental build and in a v0.x preview state. I’ve taken care to attempt to address the lethal trifecta: the main agent loops&#x2F;sessions never ingest untrusted DOM code or possess low level navigation tools. It delegates those tasks to dedicated web runners with no wider tooling or secrets access that return summarized results. Both the DOM and the summarized results are bracketed as untrusted, meaning two stacked prompt injection escapes are needed. All egress goes through a central module that has a customizable deny list, and only models calls to designated allowed endpoints are possible. See more in the docs, site, and the code itself. Ultimately, use at your own risk.<p>Today anthropic, open router, local ollama, and even an experimental WebGPU instance of Gemma are supported.<p>Honest limitations: Chrome store and AMO are still pending until it can get more eyeballs and live usage. Just loading unpacked from GitHub is the easiest way to go, and as a bonus makes it easy to audit thanks to no build. Linux on wasm depends on the Cheerpx engine, which is not open source and has restrictions for commercial use. That may be a good reason to reassess it compared to alternatives, but it’s also the most performant and looks closest to implementing 64bit support.<p>Poke around, use it, critique it, and have fun.

Show HN: TikZ Editor – WYSIWYG editor for figures in LaTeX Hi all! TikZ is a widely-used LaTeX package for drawing figures in papers. It uses commands like \draw[-&gt;] (0,0) -- (1,2); to draw lines, shapes, text, etc. Academics usually code up their figures by hand, so there is lots of twiddling around with the coordinates and recompiling until things look nice. I guess it’s a bit like SVG, but it’s more code than markup, for example it has loops with \foreach.<p>I built an open-source WYSIWYG TikZ editor (available for web and desktop) that allows you to edit your TikZ source code visually by dragging and resizing elements. It simultaneously shows the source code and the rendered figure, and lets you edit either one while the two views stay in sync. I’m not aware of any other editors that are simultaneously source editors and WYSIWYG (even for editing SVG or HTML), and I’m quite pleased with how well the combination works.<p>The way the app is implemented is by parsing the TikZ code, and at all times keeping track of the exact source location of each object. Thereby, when a user drags an element to a new position, the app can override just the numbers in the coordinate without changing anything else in the code (such as line breaks or indentation).<p>This approach essentially required reimplementing a large fraction of TikZ, which is the kind of task that no human would ever want to do. I think building software that doesn’t exist yet because it would be impossibly tedious to code up is one of the great new possibilities thanks to coding agents, and it’s worth brainstorming for other examples. (This app was built almost entirely by Codex.)<p>Implementing the app came with lots of fun side quests, including building converters from SVG &#x2F; pptx &#x2F; ipe to TikZ, re-implementing the LaTeX hyphenation and line-breaking algorithm to support multi-line nodes, and making a color picker that uses the red!20!black color mixing notation used in LaTeX papers.

Show HN: Transformer Primitives – A visual explainer you can send to anyone I have had a few conversations in the past year with non-technical folks (traditional finance types, consultants) who asked for a simple explainer on how GPTs work. These people generally have the horsepower to grasp new concepts quickly but don&#x27;t necessarily have the math background to dig in super deeply and want a stronger framework than &quot;they predict words&quot;.<p>I had generally been pointing them to Karpathy, Illustrated Transformer, and other youtube content but I figured something more hands on without the math requirement might be better at explaining questions like &quot;why matrix multiplication?&quot;.<p>I built this as a simple 60-70 minute interactive course with some demos I could send along to them. I&#x27;ve found it effective as the initial foray into understanding jargon and concepts before they graduate to Karpathy et al.<p>It&#x27;s free, no signup required. I would love feedback on where it might lose a non-technical reader, and where someone who does this for a living thinks I have simplified past the point of being correct.<p>understandgpt.xyz

MSG Made Dossier on Activists Who Opposed Facial Recognition

Israel targeted Gaza children resulting in genocide, UN inquiry says

Show HN: Shumai – open-source Frame.io alternative for creative work Shumai is an open source platform for uploading creative files, managing projects, collecting precise feedback, sharing work, and collaborating with AI agents, all in one simple creative-first workspace.<p>I’ve always liked the product design of Frame.io, and I wanted to build an alternative that feels just as polished, while being open source and easy to self host.<p>You can deploy Shumai with docker compose in just a few minutes. It can also be installed from npm, though you&#x27;ll need to provide your own PostgreSQL instance with the pgvector extension installed.<p>For larger deployments, Shumai supports distributed processing via Temporal, allowing resource intensive tasks such as transcoding to be scaled independently.<p>It&#x27;s still early in development, feel free to try and share any feedback.<p>Demo: <a href="https:&#x2F;&#x2F;staging.shumai.one" rel="nofollow">https:&#x2F;&#x2F;staging.shumai.one</a>

Show HN: Neural Particle Automata Neural CAs model self-organizing pattern formation on grids. Now the grid is gone. Each cell is an agentic particle that can move freely in space and change its state.<p>While each particle follows a simple shared rule, many together can grow complex morphologies or form intricate patterns. The resulting particle system as a whole can regenerate from damage and exhibits surprising emergent behavior.<p>Try cutting the lizard and watch it heal itself!

Show HN: Ideate a trading strategy with an Ex-Citadel Trader