https://www.noisetable.xyz/ - a collection of chance-based audio 'channels' in a VCR inspired interface
https://concrete95.net/ - a musique concrète web app that's made to look like windows 95. Pulls random audio from freesound.org and loops a small section, you can also layer synth pads or melodic synth loops. I'm often able to get some really pleasant background ambience.
https://windso.me - a sample-based step sequencer that doesn't let you choose the sample that's loaded, kinda fun, still needs a lot of work!
- sandvault https://www.codeofhonor.com/projects/sandvault/ runs agents in a separate macOS user account, hardened with sandbox-exec. It also supports headless browsing and iOS Simulator from inside the sandbox for testing web and iOS apps.
- clodpod https://www.codeofhonor.com/projects/clodpod/ agents run inside a macOS VM.
- git-multi-hook https://www.codeofhonor.com/projects/git-multi-hook/ git only allows one script per hook event; this is a dispatcher that discovers and runs every script in a hooks dir, in parallel, for both global and repo hooks.
- TubeGate https://www.codeofhonor.com/projects/tubegate/ Chrome extension to block YouTube videos based on keywords (like “sponsored”).
- push10k https://www.codeofhonor.com/projects/push10k/ iOS app to track my progress toward 10,000 push-ups.
My blog is AI-coded: Zola static site, Sveltia CMS, Cloudflare Pages/Workers, with GitHub Actions handling builds and syndication. https://www.codeofhonor.com
At the same time, I moved from Chrome back to Firefox, and Gemini was great at finding equivalents for my most-used extensions -- and, when none existed, to write my own. It's also been really useful for customization/"ricing".
More recently, I got into Quod Libet as a primary music library manager, and both Gemini and Claude have been fantastic at helping me build custom plugins that make it do exactly what I want it to. Scripts to automagically download tracks with metadata and synced lyrics, a lyrics sidebar that highlights lines as you listen and lets you click to jump to a specific line, a bookmark button that lets you mark your favorite section of a track for easy browsing later. Next chance I get is something that enables lyrics search across the entire library (a feature I was already able to build for the Stremio desktop player -- it's so cool to be able to search for a line in a TV episode or movie and jump straight to it).
I made a scraper that searches through all the news and finds anything 'war' related. It then summarizes it and give me that new in the Star Wars crawl with music.
Plus after each run you get screen recordings with console logs, network requests, HARs, and Playwright traces so you can inspect exactly what the agent did :)
https://github.com/wizenheimer/canary
P.S. I attempted to do a Show HN but got flagged for some reason
I am the only user. Sometimes it's the process that matters, and exercising your brain is important too. I get that there is a lot of existential dread around AI taking our jerbs, and excluding humans from the process of creative work, but... you can still just write code, just for the personal satisfaction.
So basically a way to just go on an hour long walk with myself, spit everything from the top of my dome stream of consciousness style, and then have Claude structure whatever I said.
It's nice to have something that structures my thoughts by just thinking out loud.
I vibecoded it (it's approaching 20K lines including tests). It works quite well but there are some bugs, so will have to do some actual engineering. But the UX is working quite well.
- https://github.com/exlee/pikchr_pl - DiagramIDE (diagram amiga-style workbench where you can script Pikchr diagrams using TCL, Prolog, Pikchr or - recently mruby). Note: you need to navigate to actual crate for description. There are binaries built in case someone wants to try it.
- https://svg.axk.sh - semi manual SVG fitter so that I can easier vectorize AI-generated pseudovector images (who doesn't like 30kb SVG versus 1.8mb PNG?!)
- https://github.com/exlee/rik - this one makes me laugh every time I use it - it's an AI harness with text editor as an UI (i.e. it reacts to comment strings) - I gave it personality so it makes wacky comments but other than that it's very constrained agent (limited edition ;))
These are not vibe coded but AI made it much easier to slide through major friction points (e.g. for SVG fitter I really didn't want to reinvent fitting algorithms)
Also used AI to design an online store (I'm not a front end dev). It's amazing to see my wife (non-technical background) tweak the web-site using claude code.
End result: an online store where we sell jewelry pieces that actually are lambda-diagrams (Tromp diagrams) that compute Graham's number, or of the Y-combinator (well, technically it's just a fixed point operator, one beta reduction away from the Y-combinator. But Y-combinator was not aesthetically pleasing from a jewelry point of view)
store: Built a tool to help design cs/science inspired jewelry in CAD. I wrote a DSL to describe the jewelry, and had an LLM write the interpreter to generate a CAD file using cadquery (note to self: LLM suck at 3d reasoning). I would not have had the time to do this without AI.
Also used AI to design an online store (I'm not a front end dev). It's amazing to see my wife (non-technical background) tweak the web-site using claude code.
End result: an online store where we sell jewelry pieces that actually are lambda-diagrams (Tromp diagrams) that compute Graham's number, or of the Y-combinator (well, technically it's just a fixed point operator, one beta reduction away from the Y-combinator. But Y-combinator was not aesthetically pleasing from a jewelry point of view)
store: https://studio-galois.com
Hister is a full text indexer for websites and local files which automatically saves all the visited pages rendered by your browser. It provides a flexible web (and terminal) search interface with offline result previews & detailed query language to explore collected content or quickly fall back to traditional search engines.
It can provide a privacy-respecting search experience for serving "recall" type searches where users retrieve previously visited content, but falls short in "discovery" type searches (yet).
(Edit: forgotten in first edition) A cookbook to store the recipes my family likes to cook so I can eventually break up with Pinterest: https://github.com/vtbassmatt/Cookbook
A data extraction pipeline and search engine for a new card game called Mood Swings: https://moodswingsdata.github.io and https://moodswingsdata.github.io/feelings.
An app to let my friends and me build a Magic: the Gathering cube iteratively together: https://github.com/vtbassmatt/popcorn-cube
A custom wiki engine for a family of podcasts I enjoy: https://github.com/vtbassmatt/totalus-wikium
A systemd log viewer for the web: https://github.com/vtbassmatt/djournal
Could I have done this myself? Of course. Would I have tho? Prob not.
This kind of simple, hyper-specific bespoke utility is the perfect thing for vibe coding IMO.
With Tailscale, you can basically point a domain to the FQDN of a machine you’re sharing with people and the domain will simply work for them (and only them)[1]. But for it to work without them having to know or specify the specific application port, you have to grant them access to 443 (and 80) in your Tailscale ACL for that specific host.
So yeah, now immich.familydomain.com works without family members having to remember the specific port. BUT, serveradmin.familydomain.com (another app on the same host) will ALSO be accessible to them (from a networking POV). We opened port 443/80 for that host after all.
I took a few hours with Claude back in January (?) and we wrote a tiny Go authorization gateway which basically consults both Tailscale’s public API and Tailscale’s `localapi` and returns the appropriate response to Caddy based on the requesting user’s actual allowed ports.
So now I can share different apps (subdomains) with different people without forfeiting access controls, all driven by Tailscale’s policy file.
(I hope I didn’t mess up the crux of it, pretty late here)
Edit: why not (something like) Authentik? Quoting from the draft:
I contemplated this for a bit but one thing I kept thinking about is:
_They're already logged in to Tailscale_
Why do I need to install and configure another full-blown app dedicated to doing something Tailscale already did? Why have the users go through two hops of authentication?”
I've always found it annoying that browsers autoplay animated images, and there still isn't a built-in way to control that behavior.
The extension shows the first frame and adds a play/stop button directly on the image.
What started as a personal utility ended up being published on the Firefox and Chrome extension stores. It's still a small project, but it solved a problem I had every day.
Firefox: https://addons.mozilla.org/en-US/firefox/addon/gif-control/
Chrome: https://chromewebstore.google.com/detail/gif-control/nhoihin...
Bug reports and feedback are very welcome.
Disclaimer: I previously posted a Show HN about it, but it didn't get traction [1].
I probably would not have bothered to allocate time to this pre ai, the juice wasnt really worth the squeeze. But I approached it with an initial amusing naivete about it being 'super simple'. As is almost always the case with software theres a reasonable amount of hidden complexity. But I have been using it as a sort of learning proving ground for how to work with agentic development. For example I got to a point where claude wouldnt implement properly and would argue with me about changes because it would read the current/old docs in the repo and get confused about reality. So right now I'm experimenting with 'canonical specs' that can only be changed modally with gates and a defined cascade from canon, to code, to docs in that order. Otherwise you end up in a weird thrash about the docs and the code disagree and which one will the agent decide to change for consistency?
Anyway, its been interesting and its v0.6 and at a point where Ive not hit a sharp edge dogfooding for a while and some beta testers would be valuable. Right now you have to manually wire it into your stack, once some others have kicked the tyres hard enough I will make some pr's to the popular tools to consider integrating it.
Later I find the technique I developed to instantly load arbitrary large CSV files can be generalized to work with any format, with an incremental parser combinator. This means the tool can read from not only files, but any stream-oriented sources like a pipe.
I did used AI in development but it didn't speed up the process very much, as I found a lot of time was spent on the deign of an intuitive and consistent UX. The project is still not very production ready, but in case anyone is interested: https://github.com/Verticalysis/Hitomi
A self hosted web archiving tool with support for extendible processing pipelines (eg. extract article -> translate -> summarize -> generate tags, download video -> split audio track -> transcribe -> summarize), which led me to make a managed chromium browser with extensions and warc support for archiving, and a RSS feed synthesizer (take random article listing page that doesn't have RSS and generate a feed for it) so that I can plug it into my archiver. An active learning loop for a model to clean up articles by removing junk like native ads and sponsored blocks.
A tabbed terminal with project management features like launching the database, app server, and claude code in different tabs with one click, and split browser/terminal panes (eg. opening a browser automatically at the correct URL when the terminal reads http://localhost:4000/).
A modular MCP server with a MCP proxy and OAuth2 dcr so that I can easily add new random ideas for MCP servers in a few minutes with Claude and deploy them such that it's available to Claude by refreshing the tool list.
A small tool to render Claude conversations so that I can link to them from my obsidian vault with something like convo://claude-code/-home-jfim-projects-foo/<guide>
And overall just deploying docker containers for my self hosted setup
Most of it is on GitHub, in various states of readiness.
https://gitlab.com/grepular/foxcage - Runs Firefox inside podman to isolate it from the host. Has some interesting features that I wanted and nothing else gave me.
https://gitlab.com/grepular/claude-sandbox - Yet another Claude sandbox. Runs it inside podman again. Has a pretty powerful proxy system for securing your credentials.
Currently working on a tool for sanitising email. Will be blogging it up at https://www.grepular.com/blog/ when it's ready for others to use. Does things like applying policies to html/svg/calendar/vcard parts to whitelist or blacklist tags/attributes/css/url schemas, clean URLs, fetch remote content at delivery time and attaching to the email to prevent tracking, pgp and smime auto encryption/decryption and a million other features.
It creates its own copy of your workdir for the agent to play in, and then you pull changes out ala git diffs or commits.
It's a MASSIVE time saver, and I use it as my daily driver.
- Automated backups of steam saves for when my kids wanted to play the same game on my account and saves get out of sync with steam cloud. The kind of thing I wouldn’t usually bother with myself.
- automated script that reruns failed GitHub actions in repositories with flakey tests ‘cause why bother fixing them? It also auto catches up branches with the main branch for the repo.
- a YNAB extension for pi (agent harness) to help with entering purchases that need to be split across categories. This is also in the telegram bot so I can use voice-to-text to explain a purchase.
- I already had some python scripts to generate pdfs of Magic: The Gathering cards for printing proxies. I had an llm extend it to make some dividers with the set names and symbols on them. Makes organizing them a lot easier in the big card boxes.
Not exactly a tool, but I also made pelohard.com which ranks the most recent Peloton classes by difficulty. Updated twice daily.
I used OpenClaw to make a health and wellness coach agent that tracks calories and alcoholic drinks and logs it to a personal dashboard. I send it photos of my meals, and it will estimate the calories and log it. It will also help me make meal decisions and give me words of encouragement.
I used this HAM dashboard git repo to create a bespoke dashboard of different video and weather feeds from my area: https://github.com/VA3HDL/hamdashboard
I also, in the same rabbit hole, created a radio reference guide for the Sonoma County area: https://mybbor.com/petaluma-sonoma-ham-radio.html
I've spun up probably close to a dozen one-off or small websites for various little interests or projects. One of my favorites is a short domain file uploader that I can quickly host Markdown and HTML files to share with family, friends, and colleagues. It's using Caddy and running on a DO-VPS. I open sourced the code here: https://github.com/RobbyMcCullough/honeydrop
Normal browsers have built-in dev tools - this has built-in design tools. so I can visit my app, open up a surrounding canvas, pull fragments into the canvas, do some design-ish stuff, and merge it back into code. All in the same UI. It was cool enough that I'm going to release it, but for now it's very useful for myself.
Various MCPs for above.
A "remote claude code server", that gives project level overview and lets me run projects / develop on my home server rather than locally through my laptop. Supports ssh as well as a web UI (projects in a list, shell rendered using https://xtermjs.org/, with a tile overview when working on multiple projects to watch for turn ends.
Similar to above, I have a local version that auto launches a new project scoped podman container, passess through the work directory, installs CC/Codex/Grok into it and passess through the auth / config for each agent. Then dumps you into that shell with aliases to map each agent to that agent with a few special env flags to disable permission prompts, so claude = `claude --dangerously-bypass-permissions`.
An extensive MCP for Obsidian that gives agents access to use a lot of the more advanced Obsidian functionality, such as suggesting and installing plugins / configs / etc.
It's helped some of my daily productivity, but I still prefer to get my hands dirty with code most of the time rather than full prompt it.
Most of the tools I write now are bridges to various SaaS products that have APIs but no CLIs.
- CleanMyMac alternative based on opensource tools - https://github.com/p-raj/open-cleanmymac
- Standup meetings to comic generation - https://github.com/p-raj/standup-to-comics
- Configurable Pomodoro - https://github.com/p-raj/open-tomato
A few more closed source ones that aren't any close to be in a working state.
When you plug it in, the device is recognised, but press any button and it attempts to start the pairing process. Then using evtest nothing is coming through.
That^ was pretty much my prompt too, and 10 minutes later I have a working driver with systemd unit so it works through restarts. Amazing stuff!
Open Camera Control https://github.com/jcubic/open-camera-control - that allows me to control the settings of my DSLR while I'm recording myself.
Horavox - https://github.com/jcubic/horavox - A speaking clock
Mutimon - https://github.com/jcubic/mutimon - a config driven web scraper (found this post from the email sent by this tool).
ASCII-Globe - https://github.com/jcubic/ascii-globe - JavaScript library that renders a rotating earth or any map. Can be used to add animation to your website.
It feels like you have parallel conversation with an agent, except it understands all the surface. And it's granular, surgical and precise.
Usage: I'm writing a game design document of 70 pages with it, working surgically without having to worry about what page number the edit was.
It really solved my main bottleneck which is telling the AI what to do in a complete and comprehensive context.
Readme is trash but your agent will understand what to do
https://github.com/ityonemo/otzel
- Spector, a bare-bones CQRS library for elixir that composes extremely well with Otzel:
https://spector.hexdocs.pm/Spector.html
- nanodrop, an elixir library that interfaces with nanodrop spectrophotometers:
https://nanodrop.hexdocs.pm/Nanodrop.html
- opengenepool, a web-based dna editing tool (technically I started this one 11 years ago):
https://opengenepool.vidalalabs.com/ https://github.com/vidala-labs/opengenepool
- a proprietary camera library that communicates with industrial-grade digital cameras.
- a full on self-hosted lab management system (journal articles, obsidian-like notes, lab notebook, kanban, embedded dna editor, LIMS), obviously this builds on all of the above items
I also built a kanban board with agent integration and context management, with a vs code extension to go with it (also helps with git worktrees too): https://www.agentkanban.io
There is AS Notes - an Obsidian / Logseq / Roam alternative for use a s a VS code extension (is designed for use behind corporate firewalls, git friendly): https://www.asnotes.io
Also NumeroMoney: https://www.numeromoney.com - For personal finance spending analysis and budgeting.
AI has been a great 'exoskeleton' for me. I fortunately had some good infrastructure and solid application base templates from before AI 'got gud' and so building on these has been the best of both worlds - a solid base and improved speed of development.
- https://blunders.ai : Chess improvement app
- https://fretwork.ai : Freelancer management app (CRM/Billing/etc)
- https://validity.ai : Provide agents the ability to check the UI code it made (w/out needing to run through your full app)
- Save money on groceries + meal planning. This has probably saved hundreds if not $1k+ for our household at this point (some details here: https://x.com/ryanlanciaux/status/2063604299590939042)
- Orchestration / Starter Kit / Chat : Tool to help me manage multiple agent sessions at once. Some details on this one here https://x.com/ryanlanciaux/status/2063976049537417408
;; Written by Claude 2026-06-06
(defun my/org-to-markdown-clipboard ()
"Export org region (or buffer) to Markdown and copy to clipboard.
With no active region, exports the whole buffer."
(interactive)
(require 'ox-md)
(let* ((text (if (use-region-p)
(buffer-substring-no-properties (region-beginning) (region-end))
(buffer-substring-no-properties (point-min) (point-max))))
(md (org-export-string-as text 'md t '(:with-toc nil
:with-author nil
:with-date nil
:with-title nil))))
(kill-new md)
(message "Markdown copied (%d chars)" (length md))))
(with-eval-after-load 'org
(define-key org-mode-map (kbd "C-c m") #'my/org-to-markdown-clipboard))I also build a ton of household glue stuff; I was never really passionate enough about the whole "homeserver" thing to spend the effort in going beyond basic video recording for my security system, but now I have all of my local-only home automation stuff wired together, mostly into HomeKit, and have been able to ditch a ton of cloud services.
Ruby on Rails: A volunteer 'jobs' board for OSS projects (ironically): https://ossvolunteers.com
JavaScript (client side+Cloudflare Worker): A map showing stop-level usage of OneBusAway across the Puget Sound region, updated daily: https://opentransitsoftwarefoundation.org/onebusaway/visuali...
Swift/iOS: An app that helps me track how much time I spend in the five heart rate zones, so that I can better focus my workouts. I'm working on version 2 right now, which will take advantage of some new features in iOS 27 and has an all new UI. https://www.zone2.app
I also wrote an honeypot that emulate an Ollama instance: beside the attackers, it's funny how many people are looking for free inference. Somebody from Brazil try to use my honeypot to write to chapters of a book about traditional magic rituals. My next step is to extract the data collected by this tool to extract IoC and malicious prompts and share them with the community.
In the same scope I wrote also an Ollama scanner: it fetch from Shodan the open Ollama instances, verify that they are reachable and check if they are real sending a dummy query.
https://github.com/Opfour/coeus-ci - Named after Coeus (SEE-us), the 100 eyed Greek Titan of the inquisitive mind — whose name literally means "querying." CI stands for Competitive Intelligence. A business intelligence OSINT tool that builds company profiles from free public data sources. Give it a domain — get back a scored report covering stability, growth, tech maturity, financial health, security posture, and transparency.
https://github.com/Opfour/op4 - Op4 is a terminal-based encrypted messaging application written in Rust. It provides end-to-end encrypted private messaging with post-quantum cryptography, routed entirely through the Tor network so that neither the content of your messages nor your IP address is exposed to anyone — not even the person you are talking to.
Spartan (Private) - Open-source safety app for women. Community-based emergency response.
AATR - (Private) Catering unified platform: events dashboard, pack lists, and staff management
https://github.com/Opfour/thelinuxreport.com - Linux news and information aggregator.
God I love this stuff!
(edit: I also have about 6 more projects I am working on locally not yet uploaded to GitHub)
Dont have the code up for sharing but I documented xeil (along with a few other tools) in my blog: https://paul.mou.dev/posts/2026-04-28-software-for-one/#xeil
And yes, there are lots of cows. The longer you stay on distracting sites, the more cows appear.
The idea is to make distracting sites less appealing, without using a black-and-white site blocker, which you can easily disable.
Chrome: https://chromewebstore.google.com/detail/mooblock/eanbagjehd...
Firefox: https://addons.mozilla.org/en-US/firefox/addon/mooblock/
Sometimes I daydream that the end goal demo would be to set one up in a VM and let a sort of ransomware lose in it, and see if the agent can identify what’s going on and react in time.
But for now I’m fighting Apple notarization to enable local notifications on macOS.
Most recently, though, is a basic python CLI/Flask app that makes it point-click to manipulate the route table and dnsmasq settings on a raspberry pi. eth0 on the pi goes to LAN and put a USB ethernet attached to a switch and you now have a pretty powerful IoT/Untrusted device inspection environment.
One click to change the DHCP settings offered to the device, run a pcap, allow/deny the device access to certain services or NAT them elsewhere as needed. Straight forward DNS adjustments that can be applied per device and now that MITM-Proxy has a python API... it's pretty straight forward to also peep inside of SSL protected things.
Or were you only asking about digital tools? I haven't really made any of those. Making physical tools feels much more satisfying these days.
And instead of turn-based, you have to hike to/visit a physical location, fill in a code you found there to get bombs for the game. Or do a quiz and get the answers right.
Great fun so far!
https://github.com/pkhamre/opencode-docker
Also made this minimalist carousel generator after seeing a carousel I really liked at LinkedIn.
https://github.com/egorelik93/Doom-Emacs-Config/blob/master/...
1. Comprehensive tool for auto expense logging and management, expense trend analysis, budget allocation, expense divider during group spend, report generator etc 2. DashCam app for simultaneously detecting threat and recording video. 3. Stock market portfolio management. A comprehensive tool which takes stock market investments as input, analysis the investments, provides a complete analysis, trends, suggestions etc 4. Fitness app. Records calories gained/burned,physical activities, all health parameters like Bp, Spo2, Blood sugar, Heartrate, Weight, Allergies and other synptoms. Analysis the health trends and provide suggestions regarding food, exercise and other health related anamolies. 5. AI learning app series ((13 apps to help learning AI from scratch) 6. Private chat app using Bluetooth communication 7.My own versions of Doc scanner, phone tracker apps 8. Health app which analyse tongue, eyes, and any symptoms to suggest the possible health issue and the remedy 9. JEE preparation CBT app 10. Electronic circuit builder and simulator app. (All basic circuits using AC/DC power, resistor, capacitor, transistor, diode, led, zener diode, switch etc can be created and run, it supports multimeter and CRO tools to measure different current/voltage and watch the waveforms)
Lot other things are in pipeline and will post once i complete them.
* plugin for Logic Pro to simulate how a mix will sound on my macbook and phone (I captured real impulse responses for that, sounds very close)
* an app for spaced repetition for guitar video lessons / their parts (no idea why platforms like truefire don’t have this feature)
* workout planning/tracking app
* an app to create impulse responses for acoustic guitar, to make it sound good live
OneNote to markdown/obsidian canvas converter. It did that using interop api to read the actual XML of the onenote files.
Work time tracker as 1px line on edge of monitor. Shows thin line at the edge of the display which fills up based on what i am doing.
Plaintext bookmark chrome extension that save links to local markdown file, Dynalist, Workflowy, Github Gist and import export between them. Was originally for Dynalist when AI couldn't do much 2-3 years ago. Recently added these other end points. [2]
A heart rate monitor with finger on camera. It's bit crappy though. Had to make it because many trackers, including google fit, couldn't detect 200bpm. https://github.com/SMUsamaShah/heart-rate
[1]: https://github.com/SMUsamaShah/LookupChatGPT/tree/claude/fix...
This one generates maps from OpenStreetMap data + some custom curated info in YAML: https://github.com/c0nsumer/trailmaps.app-map-generator
This one converts a basic chunk of OpenStreetMap data to an SVG so I can mark it up (by hand) in Adobe Illustrator to make specifically-styled print/PDF maps, such as what get installed at trailheads: https://github.com/c0nsumer/osm_to_ai
This one takes GPS recorded rides and builds custom/personal heatmaps serving up the map tiles so I can use them in map editing software: https://github.com/c0nsumer/local-heatmap-tile-server
And all of this has been put together to make the custom, local, specific-use-case maps that are at https://trailmaps.app (which, via local curation, are overall better mobile/online maps than many of the bigger auto-generated systems such as Trailforks, Gaia, RideWithGPS, etc, for visualizing local systems).
It's neat stuff where I understand all the inputs, outputs, and how most of it works, but AI tooling (Claude, mostly) has allowed me to bolt it together much faster than I would have writing it myself.
Then, I got bored because they seem to be bad beyond certain complexities. But around Christmas last year, things improved a lot, and I’m getting confident building real ones. In the mid of these, I also got an offer to work with the [pre]sales/GTM team of a large company. So, I have been building working prototypes of bits and pieces of boring enterprise business around the world and have been racking in, if not billions, but very close to it in sales pipeline in about two months. (I did that 1B token in a week thing.) I’m sure the business and sales team will be able to convert 20% to 30% of that in the coming months. I also pitch and presented my work directly to customers and they are coming for more. So, it has been fun.
Lastly, I really wanted to scratch an itch I had for a while—build a Static Site Builder. But I ended up buildig it to be a documentation tool for TypeScript and JavaScript. It can be used as an auto-detect and let it build alongside front-end projects, make it a hybrid so you can add human-written documentation too or just build a Jekyll-esque static site.
O’Vellum is a 3-in-1 documentation tool. https://ovellum.oss.oinam.com
The work silencer/breaker is at https://void.oinam.com
The bubble popper that I re-created which my daughter loves https://brajeshwar.com/2025/bubble-wrap/
- classless CSS library: https://leftium.github.io/nimble.css
- HN client: https://hn.leftium.com
- local realtime streaming transcription prototype: https://rift-transcription.vercel.app
---
These projects were started without AI, but heavily augmented with coding agents:
- https://weather-sense.leftium.com
- console.log replacement: https://github.com/Leftium/gg
- Thin layer over Google forms/sheets: https://veneer.leftium.com
imping - PingPlotter-like app. They didn't have a Linux version and I'm a paying customer, so I vibe coded this one: https://github.com/zenakuten/ImPing
utcolor - text colorizer for Unreal Tournament 2004 https://github.com/zenakuten/utcolor
utquery - Unreal Tournament 2004 Game Browser tool https://github.com/zenakuten/utquery
utstatsdb - This is an old project that did not work anymore with modern php+mysql. I had claude fix it. https://github.com/zenakuten/utstatsdb
Blocks the computer for x minutes. Agents keep on working. AI doesn't need a break, but I do. And honestly, at least for me AI has made my desktop as addicting my social media feed.
https://github.com/ankitson/webby
CF pages still required too much confusing clicking around on a webpage for me. This way I can just point any little report or app at a directory and done.
There’s others that are more server shaped and tightly coupled - a pipeline to pull in all my data like Garmin, Twitter bookmarks, messages into a Postgres DB. Kind of a personal data warehouse that i can use with apps/automations, like alerting me if my sleep schedule is drifting, and a custom web interface to my Garmin data
I realistically probably would have only saved myself less than an hour of crafting those expressions if the tool had already existed (with this level of detail, there are lots of many simpler ones already for it), but I would have spent a solid 40-50 hours of bouncing between manually crafting and writing detailed instructions to direct the agents to get this tool there.
most ambitious was browser extension to automate booking reservations since captcha and timing were critical.
swamp/evaporative cooler controller that monitors rate of change of relative humidity and sends RF signal to cooler to turn water on for x seconds to keep humidity lower than just leaving set to cool all the time. Does a great job but need to work on edge cases and ultimately replacing remote with my controller.
financial models for retirement planning
pen plotter gcode creator for old 3d printer to make labels.
food monitor to track what i ate with AI analysis based on symptoms recorded as well as interactive follow up questions
inventory app for electronic parts that now is inventory for all kinds of stuff like tools/plumbing etc that uses photo/ai to fill in data.
In-browser 4x image upscaling. Vite + Onnxruntime + https://huggingface.co/Kim2091/UltraSharpV2
It took more than a year. Why? Cloud infrastructure is too expensive.
Now I want to add some sort of recommendation engine on top, to let me discover stuff I might like (I'm not into superheroes anymore).
https://tajd.github.io/cofferdam/ - is an experimental compile time time checker that can be run for typescript/rust to be able to impose architecture, design principles that can be forgotten by agents. The goal is to be able to make it possible for an agent to be reminded about design conventions where it forgets, so it reduces the amount of context that needs to be used initially to define those sorts of conventions.
That way, you can who played without ruining the result. Then watch highlights in peace!
Few tools:
1) a combination of Python scraper and Claude skill to help family members find job by matching jobs to resume, to rank best fits
2) similar to above, but for stock data and financial news to identify movers and why they moved and see if anything is interesting.
3) a couple of attempts to import EPIC medical data from hospital into an offline app. Needs more work, data export from EPIC is crappy and a mix of images, pdfs, text, HTML and .jsonb files. Not useful at the moment
4) an application that downloads stock market data to run 15-20 strategies and back testing to identify stocks that match multiple and then run sentiment analysis on news feeds. Interesting, but semi useful. Needs lot more work.
https://github.com/pettijohn/corsair-ai-workstation-performa...
I've also got a clang-repl wrapper for this codebase that is very easy to use and makes interactive programming much easier for me.
1. family tree based on wikitree format. Transcribe records, verify/edit, then incorporate them into the tree with full citations and biographies. This one is the big one. It includes a tree browser and best genealogical practices.
2. Pool Math replacement. Log pool chemistry tests to markdown files and suggest the right amount of chemicals to balance.
3. Calorie counter. Log calories to a markdown file, look up foods and amounts in online databases, sync with garmin connect for exercise calories.
All of these are written with AI but also are interacting with pi and telegram, mostly using deepseek v4 flash.
https://smartdomainfinder.com/
It uses an LLM to generate domain name alternatives that are relevant to your keywords, then checks whether any of those alternatives (in several TLDs) are available to be registered.
Warning: It's still a bit glitchy as I haven't fixed all the issues yet. It uses LLMs, but it's not a vibe-coded app itself. If it seems to be stuck while finding domain names, just refresh the page.
https://github.com/DavidSeptimus/alfred-jetbrains-launcher
Mostly, I use it to quickly open projects in cmux, but I use it for switching between git worktrees in IntelliJ too.
Currently working on a markdown search and wiki backed by Typesense, also has good web search, fetch, crawl. This will power my personal knowledge base system as an important step towards more leverage and better outcomes.
Cloak windows from screen capture! Perfect for keeping things private while screen sharing over Teams, etc.! All other tools in the Store would trigger virus warnings when I tried to install them. Some were positioned to ask for money for basic features. My app is both free with more features and a more intuitive UI and set of keyboard shortcuts!
* Image viewer that can handle really big photos + run scripts via custom keybindings + CLIP search: https://daniel.lawrence.lu/blog/2025-10-22-sriv-simple-rust-...
It still needs a bunch of polish, but I use it a few times a week.
Instant windows switcher with custom shortcuts and instant "opt+tab" and trackpad switching. Simple does exactly what I need it to do (just bypasses the slow window switching that is annoying), with no additional features or bloat.
I've also done a TUI that combines my messages from WhatsApp + messenger + discord which is pretty handy at work.
While I was doing it I needed to render those diagrams as ASCII and I was surprised there's no Python library for Mermaid to ASCII. So I wrote one: https://github.com/fasouto/termaid (https://termaid.com/)
I then used it to build other tools -- a personal time tracker and a Wesnoth-like game (both not yet published). Basically, I maintain a Markdown file with a queue of tasks, and I run a Jaiph workflow in a loop that automatically picks up tasks one by one and develops them with no human intervention.
- Custom off-brand version of Pangolin
- Dashboard with beautiful UI for parsing traefik logs with database, filters, map and various integrations and statistics
- Samsung SmartThings Volume Control for Soundbar in Windows 11 native style
- Android App with good UI which serves as remote for switching display output modes for PC for movie night / gaming night with various toggles and for remote game streaming
- Many little one-prompt apps which run in background for QoL
- Reverse Engineering with IDA became a walk in a park
* Auto-Birthday - if you have a contact in your android contacts that has both a mobile number and a birthday in their contact info, you can choose to send them an automatic "Happy Birthday" message on their birthday at a specific time. Can do it with hundreds of contacts. Doesn't use hardly any battery or resources.
* Wrecker - stupid simple "throw a ball at a tower of bricks and try to clear the board" game. High score tables. Made in unity. High battery when in use, No battery use when not playing. Will use internet for high score data.
* GeoNote - Create Geo-fences to generate a notification when you enter a location with your custom text in the notification. My wife is always telling me, "Next time we're here, remind me to only order one piece of toast" or something like that, so I make a note, it pops up the next time we're there and we're both happy. Notes are stored locally. No internet access required. Uses Geo-Fencing which is more battery-friendly than always-on GPS access.
All my apps are free, very privacy-focused and as battery-friendly as possible.
No information leaves your device (other than the high score data in Wrecker).
You have to side-load my apps though. I'm not putting them on the Google Play Store. They're so annoying to deal with! OMG
I've done a lot of little things in Emacs since. Just minor things to improve my workflows or build up Chief of Staff type information flows.
My own Alfred replacement is actually better for me, it's tailored to exactly how I want to use an app launcher / shortcut tool.
- https://github.com/artemave/artwall rotates my desktop wallpaper through random paintings (Linux, SwayWM)
- I get my TILs through https://t.me/daily_bite_sized_fun_fact
- https://t.me/tolmach_forward_bot helps me practice French reading
- https://mini-meet.artem.rocks/ was an attempt to circumvent RKN (russians) blocking video calls; not a complete success, but works for some people
- counted the number of dudes in Big Lebowski with https://github.com/artemave/super_video_grep (https://www.youtube.com/watch?v=V7EUtIbOd1w)
I made it to deal with internet censorship in the UK, where different sites have different optimal exit jurisdictions, and most sites work fine so I can avoid the extra hops where possible.
It also works well for video streaming sites with geofencing, since the geofence itself is usually implemented in the frontend rather than the CDN. So only the frontend traffic needs to be proxied, while the bulk CDN traffic doesn't need proxying at all.
Socks5 is the ideal layer for hostname-based routing, since the proxy can see DNS names without needing to sniff TLS SNI (which would be incompatible with ESNI/ECH etc.)
iirc it was basically all done in a single prompt, and I've been using it ever since. The only issue I've encountered is with WebRTC not working properly with some services. (Presumably it breaks the NAT holepunching process or something, I never diagnosed it)
Another project that isn't quite finished is a "universal" web video downloader that works by shimming the MSE APIs and remuxing the streamed segments back into a regular video file. The idea is that if you can watch it, you can save it - including but not limited to youtube videos. I started this one pre-AI but AI was a huge help with the container format wrangling.
- gcode sender and generator https://mycnc.app
- CNC simulator https://sim.mycnc.app
- Cabinet design with door/drawer designer https://cabinet.mycnc.app
Now https://voluntold.fyi exists, and I never need to remember to manually move my single $100/year "ad free signup" off an event that has already happened to one that is coming next week.
Tuber[0] - this is my favorite, use it multiple times a week. It's just a little CLI wrapper around yt-dlp for my most common use cases - downloading the video, or the audio, or the subs. And then, if you've got the Claude CLI installed, it can also shoot the subs through Claude for a summary. I use it all the time, it's a great little thing!
Scrapio[1] - this is really specific but I was so pleased with how it turned out. You give it a list of "hacks" ("mods" for Super Mario World) and it goes out to SMWCentral, grabs each of the patch files, and patches a clean ROM. I think I only used it twice but it was just a nice way to chew through a list of hacks and get a few ROMs ready, made quick work of something that would have otherwise taken a bunch of annoying schlep work.
Lotus Eater[2] - calling this a tool is a bit undersell, but I'm still really pleased with it. It's a fan site for jamtronica greats Lotus that scrapes Nugs.net for setlist data and lets you do some mildly interesting analysis on things like song frequency and co-occurrence. Also has a per-user "shows I've attended" thing, Setlist Bingo. It's been fun to hack on.
Lastly, less a tool, more just a toy: last week Google released their Magenta model for doing live music generation. I thought it was really neat, and it's open source, so I opened it up with Claude, and after a few passes and some extremely annoying toolchain issues, I was able to add a spectrograph which does key / chord analysis to the "Collider" app, so you get a live readout of "what the band is playing" and you can pull out your guitar or whatever and join the jam with some info at your disposal. It's the kind of thing that would have taken way too much effort to be worthwhile in the past, but with AI, it's a really neat result of a fun night of weekend hacking. See the README I added in my fork for a screenshot:[3]
[0]: https://github.com/epiccoleman/tuber
[1]: https://github.com/epiccoleman/scrapio
[2]: lotuseater.epiccoleman.com
[3]: https://github.com/epiccoleman/magenta-realtime/tree/eric-mo...
This was inspired by a karpathy tweet [0] and the prototype created using another tool of mine: The LLM Plugin Generator plugin (essentially a curated collection of plugins for simonws llm cli as a few-shot prompt)
The llm-model-gateway companion plugin lets you serve models from the LLM cli as a an openai API. This allows you to use saved consortiums in your various clients as if they where a regular model. Bringing massive parallel reasoning to any workflow.
It occured to me at some time that an collection of parallel LLMs was not really a consortium. A consortium is a group of organizations. A group of groups. To rectify this I added for actual consortiums, where each member of an llm-consortium can itself be a consortium of models. e.g.
llm consortium save cns-glm-n3 -m glm-5.1 -n 3 --arbiter mercury-2
llm consortium save cns-k2-n3 -m kimi-k2.6:3 --arbiter mercury-2
llm consortium save cns-meta-glm-k2 -m cns-k2-n3 -m cns-glm-n3 --arbiter cns-k2-n3
Yes, even the arbiter/judge can be comprised of a consortium of models, bringing parallel reasoning to the task of judging parallel reasoning chains.
Consortiums can also now contain groups of specialists. These custom user-defined expert characters address the prompt from a different perspective. And a Westworld style Attribute matrix can be randomized to inject some more entropy into the process.
[0]https://xcancel.com/karpathy/status/1870692546969735361
Some other llm plugins I vibe coded:
classifai generates labels with approximate confidence derived from logprobs
llm-alias-options saves inference parameters such as reasoning effort with a model alias. (good for setting the provider in openrouter or creating a consortium of high temperature models)
llm-prompt-json adds a --json flag to return the llm logs object (good for getting conversion_id, or reasoning output in scripts)
llm-jina adds support for all jina AI specialised models and tools like web fetching, embedding and reranking.
I looked for a decent remote keyboard app to use on the tablet, and found nothing I liked.
I ended up asking an AI to make something that served a webpage that connected back to itself via websockets. It provided a keyboard and touchpad on the webpage and forwarded events to uinput.
It works well enough on tablet and phone that I haven't got around to replacing the original keyboard.
I also built this site for educating potential customers and other privacy professionals about the increasing tempo of CCPA enforcement actions driving compliance: https://ccpa.world/enforcement
I could have probably coded this from scratch quicker considering that it took me two weeks to remove all of the hallucinated imaginary enforcement actions against real companies and also the citations to non-existent California law that the models kept injecting into my enforcement summaries.
Vertical RTL by default, three paper modes (washi/sumi/ash) Offline reading, saved-list + per-position bookmarks
* Codjiflo: A code review tool inspired by Microsoft's CodeFlow: https://codjiflo.net
* A virtual replica of a digital readout (DRO) for operating a CNC machine like a manual mill: https://el400.vza.net
* Reverse engineered CNC pendant integration with CNCjs also for operating a CNC machine like a manual mill: https://github.com/pedropaulovc/cncjs-pendant-whb04b-6
* A 'docker compose' to provision email, chat and documents for human-AI hybrid teams where you can take over AI's agent's credentials temporarily: https://github.com/vezzadev/roster
* The CNC stuff will come handy for a bigger project I have to create a 1:1 replica of Albert Michelson's harmonic analyzer: https://github.com/pedropaulovc/harmonic-analyzer
* Reverse engineered Hik-Connect P2P CCTV protocol for integration with OSS like Home Assistant and Frida: https://github.com/AlexxIT/go2rtc/issues/2289
* Some patches for different OSS projects like improvements to MCP tools, Playwright, Claude Code, etc.
A git worktree shell utility to quickly switch/manage git worktrees, and a neovim telescope picker which switches all the loaded buffers into the worktree version: https://github.com/jasonwoodland/wt
A terminal multiplexor plugin for neovim: https://github.com/jasonwoodland/terminal.nvim
A neovim session management shell utility (and telescope picker) to switch neovim sessions—no need for tmux or screen anymore: https://github.com/jasonwoodland/nvs
A NeXTSTEP-inspired menu for macos called NeXTMenus: https://github.com/jasonwoodland/NeXTMenus
A custom harness for pi agent which implements strict plan/review/implement iterative cycles and approval gating
A dynamic wallpaper utility for macOS changes the wallpaper when switching between light and dark mode — also useful for setting tiled bitmap backgrounds: https://github.com/jasonwoodland/macos-dynamic-wallpaper
A small macOS utility that dims the screen on a schedule, reminding me to rest my eyes: https://github.com/jasonwoodland/EyeSaver
Then a little calorie-radar-alert app that motivates you to stay on diet before going into a shop: https://apps.apple.com/us/app/stay-lean-calorie-navigation/i...
And to test the hypothesis if the motivation actually works with "fast flashing images" I built this experiment: https://fitfatflix.fit/
When openClaw came out I build a game like map around it to orchestrate tasks: https://memefields.com/
And not published, I build a little recommender app of activities I can do only today or only this time of the year where I currently am, so I don't miss good things.
And some other stuff I'm working on right now :)
- usagi, for tracking Claude usage in the macOS taskbar. Lots of these around, but I wanted one that wasn't buggy and constantly adding features: https://github.com/duggan/usagi
- RockstarNinja, for sharing Claude plans and sessions. Since it's a bit of a data hog signups are limited, but me and my cofounder use it all the time: https://rockstar.ninja
- TweetEmbed, for browsing an offline Twitter archive and getting self-contained HTML copy/paste-able cards: https://duggan.github.io/tweetembed/
- bewitch, a terminal based metrics collector and data visualization system: https://bewitch.dev/
Each one of these has been a great way to really push on automating build processes, see what Claude can do for automatic documentation (screen captures, etc), and trying to give a distinct visual identity. In my next project, I'm trying to de-LLMify the prose it generates by using my own blog posts and aggressively pruning and integrating into the prompt.
The biggest change has just been unlearning "cost". Stuff I've subconsciously shied away from since intuition built up over a career has given me a sense of how long things take that just isn't true anymore. Still learning!
I'm probably the only one who'll ever use it but I like it.
- create your own coloring pages - https://www.coloringsai.com
- World Cup Prediction Pool - https://www.wk-pool.com
And many more
1. A dashboard that tracks my personal metrics (github, strava, todo completion, flossing)
2. A eink display for that dashboard
3. A realtime node graph that shows a codebase (and/or its diffs) in a way that I can visualize what functions call which, and under what conditions
4. A agent that automatically fills out government forms and creates invoices for my friends brewery based on the delivery notes in their google calendar.
https://football.sensecall.co.uk/
Other parents on the team love it. The live sharing is pretty handy when some aren’t able to watch the game.
> agentmux
> Configurable tmux agent launcher. Define AI agents (or any CLI) in TOML; sessions auto-launch the correct agent, tabs are colour-coded per agent, and prefix m cycles through the list.
- https://github.com/sethdeckard/loadout : TUI and CLI for managing a personal library of Claude Code and Codex skills across your machines.
- https://github.com/sethdeckard/atlas : TUI and CLI (also works as "cd launcher") that creates a smart, automatic map of every Git repository under your projects root.
I'm using these almost daily.
I think it was (a good idea), and AI made it easier for sure.
The front camera apps always show the mirror preview. Most of us hate photos others take of us, but love our selfies. It's because we groom ourselves and find our perfect angles using the mirror preview. So it's jarring when we see the photos of us taken by others. I always wanted a camera app that showed the non-mirror preview. Surprisingly most camera apps don't have this option. So I created this app to scratch my own itch:)
Note: Photos and videos are only stored in the browser. No data is sent to any servers. You can also install this as an app from your browser since it's a progressive web app.
1) Use chatgpt pro from codex cli, opencode, claude etc as you can't get it via API. This has been the biggest boost in productivity for me as I don't have to copy and paste.
https://github.com/agentify-sh/desktop
2) A small gate to make sure any agent cannot run destructive rm -rf or git reset --hard commands, it has saved me many many times
https://github.com/agentify-sh/safeexec
3) For mac users, summarizes and speaks out loud after codex finishes a turn
I find a phone call is more likely to get through to me than a reminder or alarm, which I can ignore or forget; an ordinary reminder is not as interactive.
Claude built it all and although there's a script for it, I just set the reminders in an interactive Claude code session in the directory. (Like I'll open a claude code session there and say "using the script in this directory, call me tomorrow at 7 a.m. with the message 'dr's appointment'."
It works well for me.
https://github.com/LukaWe/espCoinWatch - espCoinWatch, an ESP8266-based Bitcoin/Crypto Ticker with Weather Support.
https://github.com/LukaWe/LocalExifGeoMap LocalExifGeoMap - privacy-focused, browser-based tool to visualize and analyze GPS data from a batch of photos. generate interactive maps, heatmaps, elevation profiles, and trip statistics. etc.
On top of that, I made a Python parser that's meant to improve upon the awesome `orgparse`: https://github.com/Idorobots/org-parser
And now I'm building a CLI for Emacs Org-Mode, mostly focused on ad-hoc querying, agenda planning, etc: https://github.com/Idorobots/org-cli
It follows a lot of the conventions of Rails which is probably why it has turned out quite well
Everything was built with AI as part of a 12-in-12 challenge I'm doing, where I'm building 12 products in 12 months. https://twelve.zamith.pt/
Quietbox (https://quietbox.zamith.pt/) - Clear spammy emails from your inbox
The biggest unblock remains the tools/scripts/skills for documentation (started with Notion, network now sits on Obsidian for read/write).
A tool to periodically sync Device 42 data to Netbox (work).
- small club website: https://www.kolibrinkpg.com
- ticketing system with Stripe payments and QR scanning at the door
- Instagram/media ingestion for the club site
- genealogy tool with GEDCOM import
- scripts for downloading/archiving public-domain film material
- playlist/library tooling for DJ use
- music collaboration/sync tool for Ableton projects
- normal work stuff in a much larger existing codebase
I have become a lot more strict about process after being burned a few times. Mostly: make the change small, be clear about what it is supposed to do, check the assumptions before coding, use tests/logging/manual checks as evidence, and don’t merge anything I can’t review and explain myself.
Any parents with kids that cry in the night might benefit!
- Meetup alert for meetups that match specific topics
- A daily journal that transforms entries into chibi-style cartoons
- A cashflow forecast our stupid accounting software can't do on its own
- DIY service monitoring for a ragged collection of docker containers, cron tasks, scripts and various others
Check sugar memory for the latest thing we were working on.
2. The second thing is when making changes across a large codebase agents are also dumb at figuring this out and also grep 300 things, using tons of tokens. Instead I say
Check RemembrallMCP to analyze the impact of the change.
Nothing major, and only works with my infra, but it saves me a few minutes a few times a day to just be able to check the tab, and if there's an alert load up the full stats page.
- A Chrome extension to make design comments and ASCII wireframes right on the page you're working on and feed it to your AI coding tool as a prompt https://getdesignjam.com/
- An iRacing overlay to compare the telemetry of a reference lap against your live data https://flylapsim.com/
It's been a ton of fun to be able to dabble in software thanks to AI. Especially for small personal tools like these it's so powerful
https://github.com/michaelteter/docgen : create a single text file of your entire project, with a tree and some other useful bits. This is good for dropping into an LLM or research notebook instead of giving an LLM access to your actual project folder. It also can be put in your pre-commit script so you always have one single doc you can diff from one commit to the next.
md2pdf: markdown to PDF, relying on defaults and optional config files or cli args for formatting choices (such as page margins)
md2gslides: markdown, converted into slides, and using Google Slides API to generate the doc in my Google Drive. This saves me so much effort (I teach, so I make lesson plans/presentations all the time).
get-music: TUI app that lets me search Youtube and easily queue up to download one or more of the search results. Then I take the downloaded content, split out the audio, LLM process the video title, add metadata for music, and then provide an easy command interface for local searches and playback of downloaded content.
bookmarks: TUI for slurping all the URLs from my browser, LLM-tagging each url based on the tag list I provide with the prompt and url, and lots of features for managing priority, show/hide tags, etc. This was to help me stop worrying about having a hundred tabs open. Now I can just sweep them up into my own private, encrypted (sqlite) db.
ESL-Planner: Complete web app for building class plans for teaching English (based on params, such as student age range, skill level, specific teaching language (what we want to teach), etc. It's close to being ready to productize and release as SaaS, but I built it for myself initially.
Numerous other tools plus a guide doc listing all the tools and what they do. These resources are then made available to LLMs when I'm developing, saving me (and the LLM) the time of hand-crafting the same tooling over and over.
- An analysis pipeline that takes data from D1 (storing user searches), Posthog and Clicky (analytics) to periodically generate user journey reports.
- halalcodecheck.com - a database of emulsifier codes and food ingredients. Building this with the goal to help consumer verify and find alterantives of food ingredients across US, EU / UK and Asia.
Runs entirely in the browser, no tracking, no analytics, no ads.
It's a user daemon that runs on my machine and exposes a unix socket, and then a bunch of hooks in claude, zsh, vim, etc, that report directory and commands I've run and all that, pipes it to claude Haiku for summary, and then stores context in sqlite. It also exposes that data as MCP so I can use claude to say "hey what was I doing yesterday," or any arbitrary time range.
I find that in the age of using AI agents, "Wtf was I working on yesterday" is an even harder thing to remember for me, so this helps me kind of track everything with a database that a) has AI summaries already and b) can be accessed by AI as well as a CLI.
- github clone + extras
- a stack (FILO) based task manager / TODO list
- a CAD kernel with Blender frontend (WIP)
- a minecraft mod that makes real terminal emulators in block form
- ^ that but in Godot + a terrible "game" world (WIP)
- a somewhat failed app organizing claude workspaces
- a somewhat failed attempt at a VM framework for MacOS
- a somewhat abandoned gmail clone
- a farmland pricing model + maps etc.
- partially reverse engineered VCDS device
- a likely novel fractal system I need to work on some math to publish
- NTSC transmitter/receiver in gnuradio for the artful corruption of video
- backend for iOS appstore handling of account/subscription things
- an RSS / Podcast reader
Having worked with web facing teams there are always vastly different methods any individual uses to capture their feedback. If you or anyone you know on Mac that has 100s of screenshots on their desktop, this is aimed at bridging the gap.
Clipboards are optimized with context for LLM markdown ingestion and for use in work suites like Jira/Confluence.
Still fairly early, but I've been using the tool to help build the app itself which has been an enlightening experience.
Bunch of security tools: Some are at https://diffsec.dev others:
Jira Chrome Extension to add some notes and links to dashboard pages that I wanted for ease of use.
Small application which takes a CSV and turns it into a Registered Server List for SSMS in order to keep my list of servers updated for queries across all our databases.
Honestly one of the good use cases for AI. Small low complexity scripts and tools for assistance is a great use case. I'm amazed at the folks that are doing huge monolith rewrites with Agents and such, but I've never had good results for that. Small time saving scripts gets me a much more direct return on investment.
Here is a website I made with places.js for DC area board game events. https://dmvboardgames.com/
Can be used to resize images, but the main purpose was pixelation for a game I was making.
I write a Substack about the whole thing and have a pretty comprehensive list here: https://theautomatedoperator.substack.com/p/15-ways-im-using...
All spamming with messages and news - so I made python scripts that use the apps API to get the messages (exams and changes etc) and puts it into our family telegram group chat if relevant
I just can't sit here every morning 20 minutes on my phone logging in checking everything just to find out the food plan has changed ...
A tool that checks for new movie and tv releases, looks up ratings to see if they are worth adding to my plex server (see above about services I'm running), and then finds the magnet link and downloads them. But will only do so if my VPN is connected.
A tool that allows me to quickly build out paintball fields using my STLs of bunkers that I made, and export the full field layout as a single STL for quick painting, slicing, and printing.
It's called Commonplace: https://zby.github.io/commonplace/
I use Linux on my computer and an iPhone and wasn't happy having to use a cloud to sync my data between my devices. Syncthing/Synctrain + my apps allow me to keep everything on my private network, without ever touching a server or having to self-host something like Immich.
A web harness for another open source project (CHIRP) which lets you program channels into all kinds of handheld radios (HAM).
So I made, kind of last.fm/waka-code for agents where I track (anonymous) usage per project
here is the example of my profile: https://clankerlog.ai/kodisha
Data is fully anonymous, all I collect is (agent, model, project name [can be mapped to something else to hide the true name]) and everything is opt-in by default (per project)
https://archives.fifthrevision.com/color-generator/index.htm...
I also have a local zsh autocomplete macro that let's me type things like "git rename annotated branch" and ctrl-g and it will get me the actual command. There's also a ctrl-r mode that searches my history using natural language. This is connected to a locally run ollama so my keys don't leak.
- Snubnosed mandarin app. Vibecoded anki and tinder-like character game for mandarin which allows new vocab to be added on the fly. Also accurate text to speech for tones.
- What did I learn? Tweet summarize that takes all favored tweets and assembles into weekly categories and allows deep research on certain topics.
Calibre web UI 2.0 to replace Calibre's mediocre web interface. Used for browsing ebook catalog, searching and cataloging with a simple list feature.
Project Gutenberg local mirror UI to browse my local copy of Project Gutenberg books
A couple of MCP servers for self-hosted services to give access to OpenClaw. Currently working on a daily digest that Claw will generate that includes feeds from these: what news stories were popular in my feed reader, did my baseball team win, etc.
1 year later, with no js/ts skills at all, i got 10 custom plugins, several forks where i fixed bugs and some custom adapations, dozens of scripts and snippets and what not
Now obsidian works for me like I want, and still every day I use it Im still in awe
It is a bit hard to describe until I make a video but basically I seem to have changed the fundamental unit from a document to a chunk and it is just awesome
none of which would have been possible without AI
And a MCP-powered error tracking rails engine: https://dlt.github.io/blog/posts/mcp-powered-error-tracking-...
Code: https://github.com/VadimKey/xorpingtonian
Catalog (in Russian): https://vadimkey.github.io/xorpingtonian/
During vibe coding I found that emojis are not that simple as I thought about them.
https://github.com/haydenk/homestead - another Go project, working on a better uptime dashboard that can also be the main homepage for accessing homelab resources.
I also used AI to find and create issues and milestones to for me to get the project to 1.0.0
Highlights are that it completely free, no login required and works offline (once you "downloaded"/cached some files the first time around).
A Linux dockbar with tons of applets and support for x11 and Wayland. Works on Gnome and KDE. I always wanted to write one as I have been involved with several open source ones, but it is a lot of time to go from scratch. I use it everyday, and I am enjoying it so much!
The team behind Gel got acquired by Vercel and I already experienced falling in love with a dead database (RIP RethinkDB) so I decided to fork the concept to a TypeScript port with Svelte as the UI instead of React.
It'll live at disc.sh in a few months. Early dogfooding is promising.
EDIT: Also forgot that I removed React from GraphiQL in favor of Svelte too. https://news.ycombinator.com/item?id=48044888
Just a super minimalist thing where each day is one .txt file with the newest one at the top and a lazy loading scrolling list with every note going back 5 years
Supports CTRL + F searching, backups, and a bunch of other QoL features/macros
Its kind of a revolution that with agentic coding everyone can have their own hyperspecific customized apps
The wow moment came when it wrote syntax highlighting rules for 40 languages and file formats in ~10 minutes:
https://repo.autonoma.ca/repo/treetrek/tree/HEAD/render/rule...
https://azriel.im/disposition/
The text editor has an LSP built in to guide you to what can be entered, but see the examples.
I wanted a way to have visual documentation that can show/hide detail, and be pure svg so that it can be pasted into sites without becoming pixelated.
Also supports animations to show interactions between hosts, which always gets messy when drawing a static diagram.
Would have made them without UI with a bit more elbow grease invested in web-searching for some examples, maybe even a StackOverflow question.
Generally, I'm not a fan of LLMs and their social effects.
Since I live far away from family I also added an online game so I could play with them or show them what I was making more concretely.
I've cut some jigsaw puzzles that it made, but without access to an uv printer or a laser cutter that works reliably it's been challenging to actually make them
Among many others
- Converted invoicing to Typst (from LaTeX)
- Automation of blinds
- Automation of lights
- Python library to control lights
- ML tuning library
- ML feature interaction library
- Jupyter notebook slideshow interface
- Davinci Resolve Authomation
- Arduino eink bluetooth HR monitor
- Tons of small scripts
Being proud of the result.
THAT is a real game changer LLMs allowed me, both in my professional and my casual life.
For example this:
https://github.com/yodalf/coincan.git
or this:
https://bedtimebookhelper.com/
After coming back from paternity leave, I found that my team had really leaned in to AI driven development. This project was half catching up and half attempting to solve the burn out from the repeated books my wife and I were experiencing.
I've also built a release notes app for my QA teams, a DORA metrics app, a thing to map UX journeys with Playwright, and a ton of games and stuff. AI got me back into enjoying building things again.
I am going to develop the next version TermOnHost, which will let you connect to all kinds of hosts (Mac, Windows, Linux, or any Linux embedded system) And your hosts can connect to each other.
Updating my machine used to be a dozen scattered commands: apt, fwupd, and a pile of dev tools I want kept current (node/npm, go, claude, opencode, plus binaries like kitty and lazydocker) that each update their own weird way. Now it's one Bubble Tea dashboard that checks each tool's installed version against upstream and only downloads what's actually behind, so I can see at a glance what's current and what's not. Adding a new tool is easy enough that I just toss new ones in as I start using them.
It also has a cleanup mode that hunts down all the pesky cache files that quietly eat your disk and reclaims the space in one pass, which has saved me from the "laptop full again" scramble more than once.
Currently, testing it privately and tweaking to make it awesome.
Another tool I made for myself is automated video editor (takes folder of raw photos images, and generates edited video that can be shared online). Used it among other things to edit all my GoPro raw footage laying around (hundreds of files).
- little visual web app cataloguing a small vineyard by vine vigour
- 5 min web app that helps show my daughter netball zones per position
- web app to track manual irrigation runs of 30ish taps across our property
- a calculator to cost-compare types of retaining walls by length
- a virtual card deck for testing a game idea
- scripts to help clean up an unruly Gmail inboxBc my version uses simple copy/paste rather than deeper OS integration, it works more reliably in the Claude Code terminal (has to do with active windows or cursors or something). And bc it uses local Whisper, I also find it transcribes technical words such as “git” more accurately.
Nothing technically challenging but practical for me.
After that I started writing opub.dev because even minimal success in recent oss showed me just how much has changed, and I’m worried about how expensive everything will get for maintainers.
So, now I’m trying to GIVE people compute so they can start building a helpful filter layer above their projects.
But no honestly, unfortunately most tools I did for myself are not for hobbies but something that I needed for work... like this one (https://github.com/crmkit/crmkit) most recently.
It's really removed all the pain points from my practice hours, especially when internet access is shaky.
I also built https://github.com/lnenad/difiko as AI generates a lot of code that needs a nice way to review it.
It has "levels" of compilation, with EASY mode being about as easy as Ruby, and the compiler can present you with options to get that as strict & performant as Rust/Tokio.
I'm going to need at least a month to finish all the documentation, though.
Tubenote, a free YouTube video summarization extension. Mangata, a walking app that makes it easy to take notes and photos while walking. NotebookLM Clipper, a browser extension for importing content into NotebookLM. Knock, a notification tool that sends me a Telegram alert when Claude or Codex finishes a task.
and more products are also in the works.
- A self-hosted comment system for my blog (https://github.com/karthikeyankc/discuss).
- A custom RSS reader with AI capabilities to keep tab of our competitors at work.
- A git-based CMS for my personal blog (which was also built with AI).
Automatically rename screenshots: https://github.com/amichae2/screenshot-renamer
Pugneum, my static site generator based on pug/jade. Technically made many years before LLMs, but AI is fully maintaining it now so I think it counts. It's gotten to the point I believe it's superior to markdown.
(Transcribes the audio, marks the timestamps, so you can delete a word in the transcript, and it’ll crop out that segment in the audio)
* Highlight do-follow links.
* Spoof GPS - I live in a non-english speaking country. Sometimes Google sets my location to my gps, despite having an english vpn. This is an attempt to correct that.
* Local translate, rather than sending everything to Google.
* (non browser) An SSH selection screen, so I don't need to remember the IPs
Static site generator for my blog, or at least bits of it.
https://mithraeum.studio - local first agent and editor in C, also a few models on HF (mainly jsut qwen wrapped atm but working on from scratch) https://fieldopt.dev - SaaS for dispatching jobs to the field (technicians, trades, delivery, etc.) https://github.com/zblauser/ytcli - youtube music from the terminal in zig (ps it’s free, no sub needed)
https://mediaden.ca - iOS app for storing encrypted photos/videos on storage I (the user) exclusively owns, with zero servers, zero telemetry, and a host of other privacy related features.
Already have $180 ARR prebooked (the money that I used to pay for youtube music), looking forward for more.
if anyone has links for open-source self-hosted spotify/yt music replacement, I would gladly appreciate links
A remote image viewer to see screenshots in VMs.
A simple agent harness to drive spec to verification.
A YouTube video summarizer.
https://github.com/ozten - some public repos, but the majority are private repos
Screenshot here: https://x.com/LyleMakes/status/2063784301594853657/photo/1
No code or docs was hand written for this one.
in case you forgot instruct your ai text generating tool to do so: replaces — with -, removes emojis and changes quotes to look like human-typed (even though they are not grammatically correct)
This started off as a fancy cron with webhook and became a comprehensive runtime. I have been using it for months on several systems.
The final idea is to own all my data, but I’m still on it.
Pretty happy so far
https://third-space.astride.com.au/invite/c0378a6f-b1b9-4c26...
It has less features (no OG media or title/story analysis via Bedrock) but it focuses more on the features I like/need from an HN client
- a stateless dashboard for work that collects from 6 other APIs
- a refactor of a huge function with 8-deep indentation into readable small functions
- a road trip game for my kids where you take photos of things from the car
I can once again write Clojure apps for my phone, which is fun to do by hand, unlike more conventional tools for writing Android apps.
I use it, and have given my students access to it too - they use it to help their revision.
But I dont use "AI" to make them
I use a code generator
I like to use the smallest possible "toolchain", using the least possible resources, to build software tools
Ideally I want the tools to compile quickly on underpowered hardware
Allows me to efficiently work on multiple tasks in multiple repositories concurrently.
ive made some wallets, a incredible pair of ear muffs, and a bunch of key tags.
i keep being asked if im selling anything, and when i get the next piece together of building/buying an embroidery machine that can work on leather, i might
i still dont want to buy a proper fabric cad system, so im trying to figure out a minimal version for making glasses cases for everyone's christmas gifts. its handy being able to draw inputs for claude, but its also nails-on-a-chalkboard
- I coded myself a portfolio manager to manage all the projects that I have
- secrets management tool to avoid accidental leaks of tokens by AIs
- tool for automatic creation of training/product presentation videos for web apps
- sales training app
basically trying to see what a vertically integrated agent looks like, where the agent has deep access inside a framework and it operates from within a framework, so like, instead of reading files, opening processes etc - it gets a bunch of framework specific runtime tools(logs are the easiest example)
https://charleswiltgen.github.io/TagLib-Wasm/ – Also built for myself initially, I FOSS'd it because there was nothing like Mutagen for TypeScript/JavaScript runtimes. (I don't dislike Python, but think it's a bit of a mess.) This was my first serious project to leverage LLMs for coding.
https://pwascore.com/ – Built because I wanted to quantify how bad Safari was at PWAs. Learned that, objectively, Safari is as bad as PWAs as Firefox (which is to say, not terrible, and not to blame for why PWAs continue to be mostly-irrelevant).
It wasn't perfect, but it helped me feel confident in the final result.
When you install the CLI, it (with your permission) asks to update your base agent prompt files (e.g. `~/.codex/AGENTS.md`, or `~/.Claude/CLAUDE.md`) with info about how to use the tool.
This means all your agent chats know about SDocs, and it’s nearly always your agent which invokes the tool: “Hey Claude, sdoc me a list of all my open MRs”, etc.
I did a ShowHN about it here: https://news.ycombinator.com/item?id=47777633
Will improve the read me
token speed calculator - for estimating tg/s of ai based on ram speed and model size/params this helps in comparing different hw, estimating likely speeds i will get on hardware
prompt assembler - to create prompt and context once and reuse it in different ai's, picking and choosing context in a prompt, creating agent.md etc.
dashboard builder - for viewing gsc, ga, stripe data in one place
Eg: push a button, it shows that it's working for a while, then strongly flashes when it's done (success/failure). When you have it right under the monitor, it's like a macro pad for long-running things.
This reminds me of some of the very early peripherals you'd see on the Alto and other computers. I was surprised something like this didn't seem to exist, but maybe I'm just terrible at searching.