🛠️ All DevTools

[Other] Building a Procedural Hex Map with Wave Function Collapse

[Other] Show HN: DenchClaw – Local CRM on Top of OpenClaw Hi everyone, I am Kumar, co-founder of Dench (<a href="https:&#x2F;&#x2F;denchclaw.com" rel="nofollow">https:&#x2F;&#x2F;denchclaw.com</a>). We were part of YC S24, an agentic workflow company that previously worked with sales floors automating niche enterprise tasks such as outbound calling, legal intake, etc.<p>Building consumer &#x2F; power-user software always gave me more joy than FDEing into an enterprise. It did not give me joy to manually add AI tools to a cloud harness for every small new thing, at least not as much as completely local software that is open source and has all the powers of OpenClaw (I can now talk to my CRM on Telegram!).<p>A week ago, we launched Ironclaw, an Open Source OpenClaw CRM Framework (<a href="https:&#x2F;&#x2F;x.com&#x2F;garrytan&#x2F;status&#x2F;2023518514120937672?s=20" rel="nofollow">https:&#x2F;&#x2F;x.com&#x2F;garrytan&#x2F;status&#x2F;2023518514120937672?s=20</a>) but people confused us with NearAI’s Ironclaw, so we changed our name to DenchClaw (<a href="https:&#x2F;&#x2F;denchclaw.com" rel="nofollow">https:&#x2F;&#x2F;denchclaw.com</a>).<p>OpenClaw today feels like early React: the primitive is incredibly powerful, but the patterns are still forming, and everyone is piecing together their own way to actually use it. What made React explode was the emergence of frameworks like Gatsby and Next.js that turned raw capability into something opinionated, repeatable, and easy to adopt.<p>That is how we think about DenchClaw. We are trying to make it one of the clearest, most practical, and most complete ways to use OpenClaw in the real world.<p>Demo: <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=pfACTbc3Bh4#t=43" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=pfACTbc3Bh4#t=43</a><p><pre><code> npx denchclaw </code></pre> I use DenchClaw daily for almost everything I do. It also works as a coding agent like Cursor - DenchClaw built DenchClaw. I am addicted now that I can ask it, “hey in the companies table only show me the ones who have more than 5 employees” and it updates it live than me having to manually add a filter.<p>On Dench, everything sits in a file system, the table filters, views, column toggles, calendar&#x2F;gantt views, etc, so OpenClaw can directly work with it using Dench’s CRM skill.<p>The CRM is built on top of DuckDB, the smallest, most performant and at the same time also feature rich database we could find. Thank you DuckDB team!<p>It creates a new OpenClaw profile called “dench”, and opens a new OpenClaw Gateway… that means you can run all your usual openclaw commands by just prefixing every command with `openclaw --profile dench` . It will start your gateway on port 19001 range. You will be able to access the DenchClaw frontend at localhost:3100. Once you open it on Safari, just add it to your Dock to use it as a PWA.<p>Think of it as Cursor for your Mac (also works on Linux and Windows) which is based on OpenClaw. DenchClaw has a file tree view for you to use it as an elevated finder tool to do anything on your mac. I use it to create slides, do linkedin outreach using MY browser.<p>DenchClaw finds your Chrome Profile and copies it fully into its own, so you won’t have to log in into all your websites again. DenchClaw sees what you see, does what you do. It’s an everything app, that sits locally on your mac.<p>Just ask it “hey import my notion”, “hey import everything from my hubspot”, and it will literally go into your browser, export all objects and documents and put it in its own workspace that you can use.<p>We would love you all to break it, stress test its CRM capabilities, how it streams subagents for lead enrichment, hook it into your Apollo, Gmail, Notion and everything there is. Looking forward to comments&#x2F;feedback!

[Other] 169 production-ready skills & plugins for Claude Code, OpenAI Codex, and OpenClaw — engineering, marketing, product, compliance, C-level advisory, and more. Install via /plugin marketplace.

[Other] Show HN: VS Code Agent Kanban: Task Management for the AI-Assisted Developer Agent Kanban has 4 main features:<p>GitOps &amp; team friendly kanban board integration inside VS Code Structured plan &#x2F; todo &#x2F; implement via @kanban commands Leverages your existing agent harness rather than trying to bundle a built in one .md task format provides a permanent (editable) source of truth including considerations, decisions and actions, that is resistant to context rot

[Other] Show HN: Husky hook that blocks Git push until you do your pushups

[Other] Show HN: Satellite imagery object detection using text prompts I built a browser-based tool for detecting objects in satellite imagery using vision-language models (VLMs). You draw a polygon on the map and enter a text prompt such as &quot;swimming pools&quot;, &quot;oil tanks&quot;, or &quot;buses&quot;. The system scans the selected area tile-by-tile and returns detections projected back onto the map as GeoJSON.<p>Pipeline: select area and zoom level, split the region into mercantile tiles, run each tile with the prompt through a VLM, convert predicted bounding boxes to geographic coordinates (WGS84), and render the results back on the map.<p>It works reasonably well for distinct structures in a zero-shot setting. occluded objects are still better handled by specialized detectors like YOLO models.<p>There is a public demo and no login required. I am mainly interested in feedback on detection quality, performance tradeoffs between VLMs and specialized detectors, and potential real-world use cases.

[Other] Show HN: Run 500B+ Parameter LLMs Locally on a Mac Mini Hi HN, I built OpenGraviton, an open-source AI inference engine that pushes the limits of running extremely large LLMs on consumer hardware. By combining 1.58-bit ternary quantization, dynamic sparsity with Top-K pruning and MoE routing, and mmap-based layer streaming, OpenGraviton can run models far larger than your system RAM—even on a Mac Mini. Early benchmarks: TinyLlama-1.1B drops from ~2GB (FP16) to ~0.24GB with ternary quantization. At 140B scale, models that normally require ~280GB fit within ~35GB packed. Optimized for Apple Silicon with Metal + C++ tensor unpacking, plus speculative decoding for faster generation. Check benchmarks, architecture, and details here: <a href="https:&#x2F;&#x2F;opengraviton.github.io" rel="nofollow">https:&#x2F;&#x2F;opengraviton.github.io</a> GitHub: <a href="https:&#x2F;&#x2F;github.com&#x2F;opengraviton" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;opengraviton</a> This project isn’t just about squeezing massive models onto tiny hardware—it’s about democratizing access to giant LLMs without cloud costs. Feedback, forks, and ideas are very welcome!

[CLI Tool] Show HN: Mcp2cli – One CLI for every API, 96-99% fewer tokens than native MCP Every MCP server injects its full tool schemas into context on every turn — 30 tools costs ~3,600 tokens&#x2F;turn whether the model uses them or not. Over 25 turns with 120 tools, that&#x27;s 362,000 tokens just for schemas.<p>mcp2cli turns any MCP server or OpenAPI spec into a CLI at runtime. The LLM discovers tools on demand:<p><pre><code> mcp2cli --mcp https:&#x2F;&#x2F;mcp.example.com&#x2F;sse --list # ~16 tokens&#x2F;tool mcp2cli --mcp https:&#x2F;&#x2F;mcp.example.com&#x2F;sse create-task --help # ~120 tokens, once mcp2cli --mcp https:&#x2F;&#x2F;mcp.example.com&#x2F;sse create-task --title &quot;Fix bug&quot; </code></pre> No codegen, no rebuild when the server changes. Works with any LLM — it&#x27;s just a CLI the model shells out to. Also handles OpenAPI specs (JSON&#x2F;YAML, local or remote) with the same interface.<p>Token savings are real, measured with cl100k_base: 96% for 30 tools over 15 turns, 99% for 120 tools over 25 turns.<p>It also ships as an installable skill for AI coding agents (Claude Code, Cursor, Codex): `npx skills add knowsuchagency&#x2F;mcp2cli --skill mcp2cli`<p>Inspired by Kagan Yilmaz&#x27;s CLI vs MCP analysis and CLIHub.<p><a href="https:&#x2F;&#x2F;github.com&#x2F;knowsuchagency&#x2F;mcp2cli" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;knowsuchagency&#x2F;mcp2cli</a>

[Other] Show HN: OpenMeters – A fast and free audio metering/visualization suite This has been a pet project of mine for the past ~5 months. It consists of six visualizations:<p>- Spectrogram: Implements the reassignment method for sharper time and frequency resolution. - Spectrum analyzer: A-weighted, frequency guidelines, loudest frequency tooltip. - Waveform: Colored based on frequency, loudness, or static - Oscilloscope: Tracks pitch and autocorrelates against a reference signal, provides stability. Also includes a regular zero-crossing trigger. - Stereometer &amp; Phase meter: Linear and exponential scaling modes, adjustable windows. Multi-band correlation meter. - Loudness (LUFS &amp; RMS): Implements the latest lufs standard. Adjustable timeframes.<p>Let me know what you think, and give it a star if you think it deserves as much. :] (check out the readme for a more complete enumeration of its features. Currently only available on Linux.)

[Other] Show HN: OxiMedia – Pure Rust Reconstruction of FFmpeg and OpenCV Author here. OxiMedia is a clean-room reconstruction of FFmpeg and OpenCV in pure Rust. v0.1.0, 92 crates, ~1.36M lines.<p>Key decisions: `#![forbid(unsafe_code)]` workspace-wide, patent-free codecs only (AV1&#x2F;VP9&#x2F;Opus&#x2F;FLAC -- no H.264&#x2F;H.265&#x2F;AAC ever), async on Tokio, zero C&#x2F;Fortran deps in default features, native WASM target.<p>This is v0.1.0 -- APIs are stabilized but not yet battle-tested at scale. Performance benchmarks vs FFmpeg&#x2F;rav1e&#x2F;dav1d coming soon.<p>Feedback on API design welcome, especially the filter graph and transcoding pipeline.<p>GitHub: <a href="https:&#x2F;&#x2F;github.com&#x2F;cool-japan&#x2F;oximedia" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;cool-japan&#x2F;oximedia</a>

[Other] Show HN: I built a real-time OSINT dashboard pulling 15 live global feeds Sup HN,<p>So I got tired of bouncing between Flightradar, MarineTraffic, and Twitter every time something kicked off globally, so I wrote a dashboard to aggregate it all locally. It’s called Shadowbroker.<p>I’ll admit I leaned way too hard into the &quot;movie hacker&quot; aesthetic for the UI, but the actual pipeline underneath is real. It pulls commercial&#x2F;military ADS-B, the AIS WebSocket stream (about 25,000+ ships), N2YO satellite telemetry, and GDELT conflict data into a single MapLibre instance.<p>Getting this to run without melting my browser was the hardest part. I&#x27;m running this on a laptop with an i5 and an RTX 3050, and initially, dumping 30k+ moving GeoJSON features onto the map just crashed everything. I ended up having to write pretty aggressive viewport culling, debounce the state updates, and compress the FastAPI payloads by like 90% just to make it usable.<p>My favorite part is the signal layer—it actually calculates live GPS jamming zones by aggregating the real-time navigation degradation (NAC-P) of commercial flights overhead.<p>It’s Next.js and Python. I threw a quick-start script in the releases if you just want to spin it up, but the repo is open if you want to dig into the backend.<p>Let me know if my MapLibre implementation is terrible, I&#x27;m always looking for ways to optimize the rendering.

[Other] Show HN: Lobster.js – Extended Markdown with layout blocks and footnotes Hi HN!<p>I built lobster.js which is an extended Markdown parser that renders directly in the browser — no build tool, no framework, no configuration.<p>The entire setup is a single script tag:<p><pre><code> &lt;script type=&quot;module&quot;&gt; import { loadMarkdown } from &quot;https:&#x2F;&#x2F;hacknock.github.io&#x2F;lobsterjs&#x2F;lobster.js&quot;; loadMarkdown(&quot;.&#x2F;content.md&quot;, document.getElementById(&quot;content&quot;)); &lt;&#x2F;script&gt; </code></pre> It&#x27;s particularly useful for GitHub Pages sites where you want Markdown-driven content without pulling in Jekyll or Hugo.<p>---<p>What makes it different from marked.js or markdown-it:<p>Standard parsers convert Markdown to HTML — that&#x27;s it. lobster.js adds layout primitives to the Markdown syntax itself:<p>- :::warp id defines a named content block; [~id] places it inside a silent table cell. This is how you build multi-column layouts entirely in Markdown, without touching HTML. - :::details Title renders a native &lt;details&gt;&#x2F;&lt;summary&gt; collapsible block. - :::header &#x2F; :::footer define semantic page regions. - Silent tables (~ | ... |) create borderless layout grids. - Cell merging: horizontal (\|) and vertical (\---) spans. - Image sizing: .<p>---<p>CSS-first design:<p>Every rendered element gets a predictable lbs-* class name (e.g. lbs-heading-1, lbs-table-silent). No default stylesheet is bundled — you bring your own CSS and have full control over appearance.<p>---<p>The showcase site is itself built with lobster.js. The sidebar is nav.md, and each page is a separate Markdown file loaded dynamically via ?page= query parameters — no JS router, no framework.<p>Markdown is the one format that humans and LLMs both write fluently. If you want a structured static site without a build pipeline, lobster.js lets that Markdown become a full web page — layout and all.<p>GitHub: <a href="https:&#x2F;&#x2F;github.com&#x2F;Hacknock&#x2F;lobsterjs" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;Hacknock&#x2F;lobsterjs</a> Showcase: <a href="https:&#x2F;&#x2F;hacknock.github.io&#x2F;lobsterjs-showcase&#x2F;" rel="nofollow">https:&#x2F;&#x2F;hacknock.github.io&#x2F;lobsterjs-showcase&#x2F;</a>

[Other] Show HN: Skir – like Protocol Buffer but better Why I built Skir: <a href="https:&#x2F;&#x2F;medium.com&#x2F;@gepheum&#x2F;i-spent-15-years-with-protobuf-then-i-built-skir-9cf61cc65631" rel="nofollow">https:&#x2F;&#x2F;medium.com&#x2F;@gepheum&#x2F;i-spent-15-years-with-protobuf-t...</a><p>Quick start: npx skir init<p>All the config lives in one YML file.<p>Website: <a href="https:&#x2F;&#x2F;skir.build" rel="nofollow">https:&#x2F;&#x2F;skir.build</a><p>GitHub: <a href="https:&#x2F;&#x2F;github.com&#x2F;gepheum&#x2F;skir" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;gepheum&#x2F;skir</a><p>Would love feedback especially from teams running mixed-language stacks.

[Other] Show HN: Reverse-engineering Shockwave to save early 2000s web games (Rust/WASM) Hi HN,<p>When browsers killed NPAPI plugin support, thousands of early 2000s web games were effectively lost to time. Macromedia Shockwave, in particular, was a closed-source, highly complex, and largely undocumented beast.<p>So a few friends and I spent the last two years doing something arguably unhinged: rebuilding the entire Director engine from scratch. Today, we&#x27;re finally sharing dirplayer-rs.<p>Repo: <a href="https:&#x2F;&#x2F;github.com&#x2F;igorlira&#x2F;dirplayer-rs" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;igorlira&#x2F;dirplayer-rs</a><p>We intentionally stayed quiet on this until we had undeniable proof that it was actually possible. With our latest v0.4.1 release, we have full hardware-accelerated graphics. Complex games are booting natively in the browser and are fully playable.<p>A few childhood favorites, actually playable right now:<p>Habbo Hotel: <a href="https:&#x2F;&#x2F;dirplayer.com&#x2F;habbo" rel="nofollow">https:&#x2F;&#x2F;dirplayer.com&#x2F;habbo</a><p>LEGO Junkbot: <a href="https:&#x2F;&#x2F;dirplayer.com&#x2F;junkbot" rel="nofollow">https:&#x2F;&#x2F;dirplayer.com&#x2F;junkbot</a><p>LEGO Worldbuilder: <a href="https:&#x2F;&#x2F;dirplayer.com&#x2F;worldbuilder" rel="nofollow">https:&#x2F;&#x2F;dirplayer.com&#x2F;worldbuilder</a><p>We chose Rust and WebAssembly because we wanted a native, plugin-free way to play these games without relying on ancient executables. Parsing decades-old, untrusted binary blobs and undocumented bytecode is a memory safety nightmare, making Rust an easy choice. It also gives us predictable performance with zero GC pauses, keeping frame rates smooth.<p>The biggest headache by far has been Lingo, Director&#x27;s scripting language. It&#x27;s massive. It heavily supports 3D graphics, embedded Flash content, and the worst part: Xtras. Xtras were external distributable plugins compiled from native C code. Figuring out how to make those play nice in a modern WASM environment has been a serious challenge.<p>We did, however, successfully implement the Multiuser Xtra so games can create socket connections (which is how Habbo is working!). We still have a long way to go to support full 3D and the massive ecosystem of third-party Xtras, but the foundation is solid.<p>None of this happened in isolation. We built on years of prior work from the Shockwave reverse-engineering community: folks who have been poking at Director&#x27;s internals for years, and their prior projects, tools, and research made this possible.<p>Happy to get into the weeds in the comments: the RE process, decompiling ancient binaries, the weird quirks of Director&#x27;s VM, the Rust architecture, whatever you&#x27;re curious about.<p>And genuinely: what old Shockwave game do you wish still worked? Drop it below and we&#x27;ll see if we can get it booting.

[Other] Beagle, a source code management system that stores AST trees

[CLI Tool] CLI RSS/Atom feed reader inspired by Taskwarrior, synced using Git

[API/SDK] Unofficial Python API for Google NotebookLM

[Package Manager] If It Quacks Like a Package Manager

[Other] PeppyOS: A simpler alternative to ROS 2 (now with containers support)

[IDE/Editor] Julia Snail – An Emacs Development Environment for Julia Like Clojure's Cider