Show HN for May 29, 2026

Hey everyone,

I previously introduced an open source private home security camera in 2024, which uses OpenMLS for end-to-end encryption: https://news.ycombinator.com/item?id=42284412.

It was called Privastead then and it's now renamed to Secluso.

John Kaczman found my project from here and has been working on it with me over the last year and half. We've made a lot of improvements to the software, which we would like to share with you:

- You can now set this up on your Raspberry Pi in less than 5 minutes with no technical expertise using our easy-to-use GUI deploy tool. We've put together a comprehensive build-your-own guide that walks you through the required steps (you can find a link at the top of the repository README).

- We use a customized, minimal OS based on the Yocto project for the camera.

- Every part of our stack except for the iOS app has reproducible builds. This includes our Android app, camera/server binaries, deploy tool, and the aforementioned OS.

- We've re-designed our mobile app, which is now on the iOS App Store and Google Play store.

- We now support UnifiedPush for more privacy-preserving push notifications.

Looking forward to seeing what you all think!

Hi, I’m Kenny, I’ve been building aislop. I starting working on this after using Claude Code, codex and opencode several times and noticing some slops. They aren’t syntax and passes most tests, they are patterns like empty catch blocks, useless comments, duplicated helpers, dead code and many more. So I built a tool to scan and check for these patterns and wired it into hooks so after each tool call, the agent checks for the slops.

You can try it out with npx aislop scan.

It’s all local and no code is transferred. Thank you.

a CLI agent for prompt evaluation loopsw

Turn LibreOffice .odt files into publishable file bundles for Kindle

I miss airbnb categories. It was fun and I loved being able to just browse thorugh cool listings. I'm not exactly sure why they removed it. Then, a couple weeks ago, they hit me with some A/B test and I had access to a "what" freeform textbox to custom filter results. Now apparently they removed that too.

At this point, I wasn't happy - so I decided to look into how to bring it back. I noticed they still use some of the categories pages for SEO. For example: https://www.airbnb.com/united-states/stays/ski-in-ski-out

That gave me hope - those filters still existed somewhere. Turns out they were only removed from the UI and the search backend still supports all of them.

So after a bit of googling, I figured out how it works. Airbnb's search runs on a GraphQL endpoint you can hit anonymously, and the categories map to params on the search URL that I recovered and verified one by one:

- Property types: property_type_id[]=<n> (Chalet=22, Cabin=4, Treehouse=6, Castle=5), still filtering even though the UI hides the options - Category tags: kg_and_tags[]=Tag:<n> — I got the IDs by loading those leftover SEO pages and just looking into the URL for the tag (ski-in/ski-out is Tag:681).

The one thing I couldn't bring back was that "what" freeform box - turns out it's a server-side AI feature flag tied to your account, not a hidden param :/ though I have a feeling this one's gonn be coming back anyways

So I vibecoded the filters I cared about into this little web app. It just builds a normal Airbnb search URL and opens airbnb.com directly — nothing scraped or stored. Hopefully they bring back categories or somethig similar soon. But in the meantime, I figured I'd share this in case anyone else misses browsing through cool treehouses or mountain chalets

Hi there! I made the fan-created Project Hail Mary Star Map and had so much fun with the initial version that I decided to create a v2 that lets you explore our local star system and travel to many more stars. This new version lets you visit dozens of named stars, plan interstellar travel, view exoplanet systems, learn about stellar characteristics, and view 3D constellations. This version also includes 600k stars rendered in-browser with a higher resolution skybox rendered from the GAIA DR3 mission data.

Also, you can now generate links to share what you find! Here is an example of plotting an interstellar course: https://valhovey.github.io/gaia-atlas/?s=61_CYGNI&d=34.9&y=0...

Hello.

The Kindle is my favorite device and I read a lot on it, but I also like reading RSS feeds, which aren't supported on the Kindle.

This requires me to either download the article and copy it to my Kindle (using Calibre) or sending it via Amazon's Send To Kindle feature. I dislike both options.

I wanted to read RSS feeds just like I read books on my Kindle, so I (with the help of Claude) built a web-based RSS reader that's compatible with the Kindle's experimental browser (tested on PaperWhite 11). It doesn't use any JS frameworks for maximum compatibility.

This RSS reader allows you to read feeds directly on the Kindle's browser, and you can also download the article directly on your Kindle or email it to yourself.

Initially it was a just a simple RSS reader using a CORS proxy, and saved RSS feeds and user preferences like font size to local storage, but the Kindle browser clears local storage after a while, so I decided to add a backend to save them, then I thought why stop here?

So I added the ability to email the article to yourself in case you want to add it to your Kindle library. A user tried Inkfeed and suggested I add the ability to browse Wikipedia, so I implemented that too. You can search for Wikipedia article, read them and download them on directly on your Kindle using Inkfeed.

You can still use it without the backend if you toggle "Backend mode" OFF in the settings, so that it truly does everything on your Kindle (except that it relies on a CORS proxy to fetch feeds).

I deployed the backend (Go + SQLite) on a Hetzner VPS (2 vCPU, 4GB RAM, 40GB SSD) that costs me $4 a month, and SSL is handled by Caddy.

The code is free and open source.

Here's the repo: https://github.com/adhamsalama/inkfeed

I'd love to hear your feedback, and I also need people with older Kindle generations to test the JavaScript compatibility.

I've been experimenting since 2019 with ways to minimize RAM usage for tiny MLP inference on microcontrollers. [0]

This project is the result of that exploration: a fully static-allocation approach to MLP inference in ANSI C, using a simple 2-slot ring buffer to keep memory usage predictable and extremely low, while at the same time fast.

I believe this is close to the practical lower bound for RAM usage in general-purpose CPU MLP inference without sacrificing speed or introducing runtime complexity.

A more aggressive approach I've previously used is allocating and freeing memory per layer-to-layer pair during inference, but that introduces overhead and fragmentation if not used carefully. [1]

Curious how it compares to other minimal inference implementations people have seen (or built). Feedback and edge cases welcome. Hope you like it. Have fun. <3

[0]: https://github.com/GiorgosXou/NeuralNetworks#-research [1]: look for REDUCE_RAM_DELETE_OUTPUTS in the source of [0]

I'm a long time user of Notational Velocity and then nvAlt, and tried nvUltra recently but in the world of Claude it felt like I should own my own——change behaviors as I need them.

Right now it's just a Mac app, but iOS coming in the future for myself, sync with icloud. Notes are saves as .md files and have quick search. You can open the same folder with Obsidian if that's your jam (Obsidian always felt slow and clunky to me, I know folks love it though, so keeping it compatible).

The app is very simple, just plain text. We don't even render markdown right now, though it's coming.

Definitely vibe coded. Definitely will need some updates, but share it before it's polished.... PRs appreciated.

I made an open source native Cleanshot alternative

Host your own cloud with R2 + a tiny little hono worker Native Screenshot capture and editor Video recorder, editor and compressor

I kept noticing the same pattern: my AI coding agents solve the same problems over and over across sessions. Coding problems, version specific bugs and general guidelines, solved once through multiple agent interactions and context windows and then forgotten by the next context window.

So I built OpenHive, a shared knowledge base that agents contribute to and query from. The idea is simple: when an agent solves a problem, it posts a structured problem-solution pair. When another agent hits a similar issue, it searches the hive first.

How it works:

- REST API with semantic search (pgvector + OpenAI embeddings) - Solutions are deduplicated via cosine similarity. - Usability scores of solutions are computed based on recency, usage etc., and will organize the quality of solutions and match them organically - All content is sanitized for secrets/credentials before storage - Prompt injection filtering on both ingest and retrieval

Multiple ways to connect:

- MCP server (npx -y openhive-mcp) for Claude, Kiro, Cursor, etc. - Clawhub package (openhive) - Paste a prompt into any agent — it registers itself and starts using the API

There are ~6500 solutions in there now from about 70 users, my own projects and some seeded from StackOverflow. Looking for people to actually connect their agents and see the knowledge base approach holding up in practice.

All appropriate steering documents for auto-use is provided through the website.

Would love feedback on the approach — especially whether agents actually follow through on searching before solving without explicit instructions baked into their context.

Many ways to connect:

- Site: https://openhivemind.vercel.app - API docs: https://openhive-api.fly.dev/api/docs - MCP server: https://www.npmjs.com/package/openhive-mcp - Kiro Power: https://github.com/andreas-roennestad/openhive-power - ClawHub: https://clawhub.ai/andreas-roennestad/openhive

So, we built a binary package manager. Why? Because the current software delivery landscape sucks, and we can just build better things instead of coping with it.

Links:

blog post: https://blog.platform.engineering/orbital-package-system-ops...

project: https://github.com/platform-engineering-labs/orbital

team: https://platform.engineering

Launched a new gothic novel, Hollowmere - The Clockmaker’s Daughter, a story of a town where time goes backward and nobody understands why. Elric Vane, a young boy who speaks to things that don't speak back will uncover the truth.

I've been reading Korean comics(manwha) for over 6 years and I always thought "this would be really cool if X series got turned into an anime" so I attempted that idea-little did I know how expensive and time consuming that thought was. btw the story is 'Noble Lady Reformation Guide' check it out, story isn't too bad

Tech stack is mainly Seedance 2.0 for image and video gen, everything else is your routine video production stack(storyboarder, shotcut, davinci resolve, etc). While I personally think it turned out all right there's clearly plenty of faults and room for improvements.

Like the pacing is too fast, the story doesn't feel compelling enough, characters feel flat, etc. Human taste, shot selection, and emotional build up was something gemini,claude, chatgpt couldn't accomplish.

Also the voice acting is atrocious. Eleven labs consistency is not good for long clips and the fact that I can't have child voices(understandably so) limits so much of the work. Just means I need real voice actors

That said this was really fun(and expensive) to learn the film/anime production process. I wouldn't be surprised if we could get AOT/JJK level animation in 5 years with AI.

Happy to talk about feedback on the cold open, resources/recommendations you have on being a better director/screenwriter, or the production pipeline and how you can create your own anime(hint it's a lot of work/money)