|
68783
|
cheese by @Fishyy | Stardance
|
Fishyy
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: cheese by @Fishyy | Stardance
|
1793749737.3513339
|
1794354537.3513339
|
|
25232
|
cheese by @893245789347053 | Stardance
|
893245789347053
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: cheese by @893245789347053 | Stardance
|
1793400911.1462953
|
1794005711.1462953
|
|
53186
|
cheekybunny's site! by @cheekybunny | Stardanc cheekybunny's site! by @cheekybunny | Stardance...
|
cheekybunny
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: cheekybunny's site! by @cheekybunny | S Title: cheekybunny's site! by @cheekybunny | Stardance...
|
1793818345.8110294
|
1794682345.8110294
|
|
53319
|
checks wether anywhere by @Sarath | Stardance
|
Sarath
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: checks wether anywhere by @Sarath | Stardan Title: checks wether anywhere by @Sarath | Stardance...
|
1793483266.435844
|
1794088066.435844
|
|
30136
|
checkmate-chess engine by @Eshan | Stardance
|
Eshan
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: checkmate-chess engine by @Eshan | Stardanc Title: checkmate-chess engine by @Eshan | Stardance...
|
1793411514.281612
|
1794016314.281612
|
|
38551
|
chax by @chaitanyasharma645 | Stardance
|
chaitanyasharma645
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: chax by @chaitanyasharma645 | Stardance
|
1793541041.782742
|
1794145841.782742
|
|
5578
|
chat™ for macOS by @e | Stardance
|
e
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: chat™ for macOS by @e | Stardance
|
1793480126.8312807
|
1794084926.8312807
|
|
50591
|
chat™ by @e | Stardance
|
e
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: chat™ by @e | Stardance
|
1793350502.8767476
|
1793955302.8767476
|
|
40081
|
chatting app by @shaxribonu | Stardance
|
shaxribonu
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: chatting app by @shaxribonu | Stardance
|
1793998424.7147028
|
1794603224.7147028
|
|
10527
|
chatbot by @shw616 | Stardance
|
shw616
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: chatbot by @shw616 | Stardance
|
1793306857.7242143
|
1793911657.7242143
|
|
44663
|
chatbot by @dhruvanshjariwala | Stardance
|
dhruvanshjariwala
|
1
|
8.0
|
8
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-08-06T17:47:04Z", "hours": 8.0, "multiplier": 1.0, "payout": 8, "blessing": "neutral", "text": "just a parody chatbot and i am proud it", "repo": "https://github.com/codemaster010pro/chat_bot", "demo": "https://chat-bot-ui-2cqb.onrender.com/"}]...
|
[{"post_id": 42096, "ts": "202 [{"post_id": 42096, "ts": "2026-08-06T17:40:30Z", "hours": 7.8158, "text": "just making parody chatbot"}]...
|
Title: chatbot by @dhruvanshjariwala | Stardance
Title: chatbot by @dhruvanshjariwala | Stardance
Devlogs:
just making parody chatbot
Ships:
just a parody chatbot and i am proud it...
|
1793918894.6185308
|
1794005294.6185308
|
|
38530
|
chat-app by @some1u | Stardance
|
some1u
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: chat-app by @some1u | Stardance
|
1793317758.4747028
|
1793922558.4747028
|
|
19730
|
chat by @shash | Stardance
|
shash
|
0
|
249.0
|
0
|
[]
|
[{"post_id": 21107, "ts": "202 [{"post_id": 21107, "ts": "2026-07-02T21:59:19Z", "hours": 4.0, "text": "devlog"}, {"post_id": 20800, "ts": "2026-07-02T14:34:02Z", "hours": 10.0, "text": "another devlog"}, {"post_id": 20448, "ts": "2026-07-02T01:24:29Z", "hours": 2.0, "text": "anoher devlog"}, {"post_id": 19999, "ts": "2026-07-01T14:05:01Z", "hours": 33.0, "text": "dev log"}, {"post_id": 18738, "ts": "2026-06-29T22:30:05Z", "hours": 136.0, "text": "longass devlog"}, {"post_id": 12096, "ts": "2026-06-19T18:58:36Z", "hours": 4.0, "text": "devlog"}, {"post_id": 11962, "ts": "2026-06-19T14:58:39Z", "hours": 19.0, "text": "devlog"}, {"post_id": 11052, "ts": "2026-06-18T02:39:41Z", "hours": 8.0, "text": "."}, {"post_id": 10598, "ts": "2026-06-17T12:54:14Z", "hours": 6.0, "text": "nextdelog?"}, {"post_id": 10421, "ts": "2026-06-17T03:46:36Z", "hours": 3.0, "text": "more devlogs"}, {"post_id": 10225, "ts": "2026-06-16T21:10:28Z", "hours": 4.0, "text": "devlog 4?"}, {"post_id": 10053, "ts": "2026-06-16T17:01:07Z", "hours": 3.0, "text": "dev log 3"}, {"post_id": 9918, "ts": "2026-06-16T13:00:47Z", "hours": 2.0, "text": "devlog 2"}, {"post_id": 9786, "ts": "2026-06-16T07:03:34Z", "hours": 15.0, "text": "chat"}]...
|
Title: chat by @shash | Stardance
Devlogs:
chat
d Title: chat by @shash | Stardance
Devlogs:
chat
devlog 2
dev log 3
devlog 4?
more devlogs
nextdelog?
.
devlog
devlog
longass devlog
dev log
anoher devlog
another devlog
devlog...
|
1793882980.325589
|
1794746980.325589
|
|
57606
|
chasis-decoder by @hazypak | Stardance
|
hazypak
|
1
|
5.81
|
0
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-09-03T13:33:19Z", "hours": 5.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "I finally completed my web app , chasis decoder , that fetches info from rest API’s about a car from its VIN number", "repo": "https://github.com/hazypak/chassis-decoder", "demo": "https://chassis-decoder.vercel.app"}]...
|
[{"post_id": 61818, "ts": "202 [{"post_id": 61818, "ts": "2026-09-04T22:11:43Z", "hours": 0.8147, "text": "Re-did the readme , and fix some errors , removed ads aswell"}, {"post_id": 60490, "ts": "2026-09-02T21:10:52Z", "hours": 0.6075, "text": "started working on cleaning up chassis decoder since the initial codebase was generated by the manus agent and looked way too generic ai-styled fired up claude code on the desktop app with the design and code review plugins enabled to help strip out the bloat and push for a clean minimalist ui that isnt overwhelming and just focuses on inputting a vin and getting telemetry specs efficiently told it to strictly preserve the ad spaces without breaking the layout or logic and to scan for hydration errors and glitches while mapping out a step-by-step plan before touching the code coming together nicely cheerios"}, {"post_id": 60379, "ts": "2026-09-02T17:51:19Z", "hours": 4.5564, "text": "this was kind of a different project for me but felt pretty similar to an old side project i did for practice where i fetched info from public nasa apis to check mars weather conditions lol . for chassis decoder i hooked up rest apis and external databases to pull reliable vehicle specs engine telemetry and chassis info i even tried adding ads to monetize it but hit a wall with google adsense because it couldnt load my sitemaps properly so gotta debug that next well im always trying out new stuff ig ,cheerios!"}]...
|
Title: chasis-decoder by @hazypak | Stardance
Dev Title: chasis-decoder by @hazypak | Stardance
Devlogs:
this was kind of a different project for me but felt pretty similar to an old side project i did for practice where i fetched info from public nasa apis to check mars weather conditions lol . for chassis decoder i hooked up rest apis and external databases to pull reliable vehicle specs engine telemetry and chassis info i even tried adding ads to monetize it but hit a wall with google adsense because it couldnt load my sitemaps properly so gotta debug that next well im always trying out new stuff ig ,cheerios!
started working on cleaning up chassis decoder since the initial codebase was generated by the manus agent and looked way too generic ai-styled fired up claude code on the desktop app with the design and code review plugins enabled to help strip out the bloat and push for a clean minimalist ui that isnt overwhelming and just focuses on inputting a vin and getting telemetry specs efficiently told it to strictly preserve the ad spaces without breaking the layout or logic and to scan for hydration errors and glitches while mapping out a step-by-step plan before touching the code coming together nicely cheerios
Re-did the readme , and fix some errors , removed ads aswell
Ships:
I finally completed my web app , chasis decoder , that fetches info from rest API’s about a car from its VIN number...
|
1793933592.5487344
|
1794019992.5487344
|
|
57544
|
chaptr by @salarmuzair | Stardance
|
salarmuzair
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: chaptr by @salarmuzair | Stardance
|
1793443856.3745499
|
1794048656.3745499
|
|
68064
|
chapter tagger by @cosmic79conqueror | Stardance
|
cosmic79conqueror
|
1
|
11.35
|
0
|
[{"ship_num": 1, "ts": "2026-1 [{"ship_num": 1, "ts": "2026-10-11T17:10:45Z", "hours": 11.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "This is my Computer Science IP Project !! the problem statement: CREATE A MENU DRIVEN PYTHON PROGRAM AND USE A BACKEND LIKE SQL TO STORE THE INFORMATION. not all the features have been implemented and are kept for future versions Here, I will be using the terminal itself to run the code and not go for any UI creation as it makes the project a lot simpler and makes me confident that i can answer any viva questions that come my way. The main idea: I am a JEE advanced aspirant. I have an error logger powered by ai agent already there. So i will use this chance to make a tool so that i can rate how proficient i am at each subject….. I will be mainely using foreign keys and primary keys to ensure smoothness.. i initially thought of creating N tables for N subjects but i was adviced against it so here i will have 4 tables to prevent clash if ever 2 people create the same subject. Requirements and how to run: You must have a mysql server installed Clone this repository by running the following command: git clone https://github.com/Abhrajit-Gogoi/cs_ip run the install.py file. This initialises the database and tables. It will ask for database name change it to you preference. If your sql username and password are different, change it in line 7 of install.py python install.py demo VIDEO: This was made for my shcool computer science ip project so I had to create a full report on it too along with this :( too much extra work uffff!!!", "repo": "https://github.com/Abhrajit-Gogoi/cs_ip", "demo": "https://github.com/Abhrajit-Gogoi/cs_ip/releases/tag/works"}]...
|
[{"post_id": 83190, "ts": "202 [{"post_id": 83190, "ts": "2026-10-27T07:36:59Z", "hours": 0.3469, "text": "made the requested changes to the ship and created a binary executable"}, {"post_id": 78796, "ts": "2026-10-11T17:05:16Z", "hours": 0.2567, "text": "done and ready to shipppp i am finally done with this!!! phew and it is ready to ship although few features are left"}, {"post_id": 78676, "ts": "2026-10-11T08:42:57Z", "hours": 1.0408, "text": "there was a mapping bug again made the changes to the chp table so as to include which all users can access which subjects"}, {"post_id": 78527, "ts": "2026-10-11T03:25:19Z", "hours": 0.8283, "text": "created a file where you can run to initialise You can run this file to like initialise all the required stuff like database and tbales……"}, {"post_id": 78029, "ts": "2026-10-09T07:36:33Z", "hours": 1.2583, "text": "added 3 funcitons now moving on to make the main execution loop but i keep running into some logical error. the loop just shuts down automatically"}, {"post_id": 77750, "ts": "2026-10-08T04:58:05Z", "hours": 3.0428, "text": "I ran into a mapping problem the original structure i decided to use had complicated mapping and i ended up entangled in it. so i went back dropped all the tables and rewrote the structure scratch up so now it is a linear mapping rather than many to one mapping"}, {"post_id": 77264, "ts": "2026-10-06T10:55:14Z", "hours": 1.7625, "text": "done!!! finally finished manually making the whole flowchart and getting it to fit on a pdf. the hard part was how branched it gets even for projects like these which are dirst simple…. I dont want to have to make a flowchart for my college projects in the futture 😭"}, {"post_id": 77226, "ts": "2026-10-06T05:30:28Z", "hours": 0.9294, "text": "created flowchart part 1 Created the actual digital version of the flowchart i sketched out in app.diagrams.net well half of it till now rest of the work is going to have to wait for timorrow……. Phew school stuff is hard work 🥹🥹"}, {"post_id": 77221, "ts": "2026-10-06T04:30:05Z", "hours": 0.4725, "text": "created the flow chart of the project on notebook I forgot to turn on the lapse until i was almost done with it most of the part was missed and this thing is so ragebaiting. “YOU HAVE TO MAKE A FLOW CHART EXPLAINING EACH AND EVERY THING INCLUDING ALL THE FUNCTIONS” 🙄🙄"}, {"post_id": 77004, "ts": "2026-10-05T10:39:19Z", "hours": 1.8117, "text": "😭 I should’ve paid mor attention in CS class I got a broken sql server which took GODDAM 2 WHOLE CS HOURS!!!! HOW IS IT SO HARD TO SET THIS UP NOT TO MENTION I HAVE TO CREATE AN ENTIRE PROJECT ON IT"}]...
|
Title: chapter tagger by @cosmic79conqueror | Star Title: chapter tagger by @cosmic79conqueror | Stardance
Devlogs:
😭 I should’ve paid mor attention in CS class I got a broken sql server which took GODDAM 2 WHOLE CS HOURS!!!! HOW IS IT SO HARD TO SET THIS UP NOT TO MENTION I HAVE TO CREATE AN ENTIRE PROJECT ON IT
created the flow chart of the project on notebook I forgot to turn on the lapse until i was almost done with it most of the part was missed and this thing is so ragebaiting. “YOU HAVE TO MAKE A FLOW CHART EXPLAINING EACH AND EVERY THING INCLUDING ALL THE FUNCTIONS” 🙄🙄
created flowchart part 1 Created the actual digital version of the flowchart i sketched out in app.diagrams.net well half of it till now rest of the work is going to have to wait for timorrow……. Phew school stuff is hard work 🥹🥹
done!!! finally finished manually making the whole flowchart and getting it to fit on a pdf. the hard part was how branched it gets even for projects like these which are dirst simple…. I dont want to have to make a flowchart for my college projects in the futture 😭
I ran into a mapping problem the original structure i decided to use had complicated mapping and i ended up entangled in it. so i went back dropped all the tables and rewrote the structure scratch up so now it is a linear mapping rather than many to one mapping
added 3 funcitons now moving on to make the main execution loop but i keep running into some logical error. the loop just shuts down automatically
created a file where you can run to initialise You can run this file to like initialise all the required stuff like database and tbales……
there was a mapping bug again made the changes to the chp table so as to include which all users can access which subjects
done and ready to shipppp i am finally done with this!!! phew and it is ready to ship although few features are left
made the requested changes to the ship and created a binary executable
Ships:
This is my Computer Science IP Project !! the problem statement: CREATE A MENU DRIVEN PYTHON PROGRAM AND USE A BACKEND LIKE SQL TO STORE THE INFORMATION. not all the features have been implemented and are kept for future versions Here, I will be using the terminal itself to run the code and not go for any UI creation as it makes the project a lot simpler and makes me confident that i can answer any viva questions that come my way. The main idea: I am a JEE advanced aspirant. I have an error logger powered by ai agent already there. So i will use this chance to make a tool so that i can rate how proficient i am at each subject….. I will be mainely using foreign keys and primary keys to ensure smoothness.. i initially thought of creating N tables for N subjects but i was adviced against it so here i will have 4 tables to prevent clash if ever 2 people create the same subject. Requirements and how to run: You must have a mysql server installed Clone this repository by running the following command: git clone https://github.com/Abhrajit-Gogoi/cs_ip run the install.py file. This initialises the database and tables. It will ask for database name change it to you preference. If your sql username and password are different, change it in line 7 of install.py python install.py demo VIDEO: This was made for my shcool computer science ip project so I had to create a full report on it too along with this :( too much extra work uffff!!!...
|
1793839939.8169882
|
1793926339.8169882
|
|
55541
|
chaox-time by @chaox | Stardance
|
chaox
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: chaox-time by @chaox | Stardance
|
1793829314.0030735
|
1794434114.0030735
|
|
52951
|
chaox-car by @chaox | Stardance
|
chaox
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: chaox-car by @chaox | Stardance
|
1793465499.584023
|
1794070299.584023
|
|
20998
|
chamber of emotions by @theeveryman | Stardance
|
theeveryman
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: chamber of emotions by @theeveryman | Stard Title: chamber of emotions by @theeveryman | Stardance...
|
1793973125.1167312
|
1794577925.1167312
|
|
8398
|
centaur by @chiron_dev8 | Stardance
|
chiron_dev8
|
2
|
19.0
|
152
|
[{"ship_num": 1, "ts": "2026-1 [{"ship_num": 1, "ts": "2026-10-31T22:05:56Z", "hours": 15.0, "multiplier": 8.31, "payout": 124, "blessing": "neutral", "text": "Intro This is my second ship for centaur , my own source-based package manager. It can install, uninstall, manage dependencies, update, sync and clean packages. Please read the changelog below for all the awesome changes i’ve made! Attention reviewers Hi, thank you for taking the time to review my code, it means a lot. This package manager is only meant for linux operating systems which means that THIS WILL NOT WORK ON WINDOWS. If you do not want to set up a linux virtual environment to test this, please skip this project. I had a couple people complain that it couldn’t run on Windows and rated my project really poorly which isn’t very fair on me. Since this project deals with system tools it can potentially corrupt packages that are already installed on your system. Please consider using a virtual environment to test this. If you do not want to set up a linux virtual environment, and you are already running linux, please condider only installing writedoc using sudo centaur install writedoc as this is a package that I made myself so won’t corrupt anything you already have on your system. Changelog Quite a lot has happened since my last ship and I have taken onboard a lot of feedback from my reviewers (thank you!): Tarballs now do not manually have to be downloaded in the install section of the package script and all the necessary data such as the source link and sha256 checksum are included in the metadata. Tarballs are automatically checked for validity by comparing it’s known official checksum and the checksum of the downloaded file. Instead of blindly running commands directly on the host system, the commands are instead run through a secure sandbox environment using bubblewrap . This allows commands on the host computer to be used to configure and compile packages but files and directories that are not needed are read-only and cannot be modified by the packages. For example, to install htop , the make and bash commands can still be run but files and directories not related to the installation of it are read-only and cannot be modified or deleted. When a package tried to add new files to the host system, the sandbox intercepts them and adds it to a staging directory instead which can then be used for different purposes after installation. After installation is finished, the new files are then copied to the host system. A manifest file is now used to track what files were added to the host system when the package was installed. The staging directory is scanned and the files that are new are added to the manifest. When a user wants to uninstall a package, all the files that are referenced in the manifest are deleted. This means that the package script creators no longer need to make a separate uninstall block. A couple new packages have been added such as htop , tmux and libevent . There are loads more changes that are slightly less interesting but have improved the performance and readability of the code :)", "repo": "https://github.com/Chiron8/centaur", "demo": "https://github.com/Chiron8/centaur/wiki/2.-Install"}, {"ship_num": 2, "ts": "2026-07-05T17:05:56Z", "hours": 4.0, "multiplier": 6.73, "payout": 28, "blessing": "neutral", "text": "I have worked on the project for a while before Stardance started. Probably around 95% of my hours are unlogged. I have made a package manager for UNIX based systems (specifically for my custom Linux based operating system, ChironOS). It can install, uninstall and update packages, manage dependencies, sync install files to the latest versions and automatically remove all unused packages. My favourite part of the project was finally seeing all the functions working together. I will still be working on this for a long long time as there is a lot I can improve on. My next goal will be to make package scripts for the whole of Linux From Scratch and Beyond LFS as this will prove the reliability of my package manager. If you want to test this please make sure you are using Linux (It will probably work on BSD but I haven’t tested it. It probably won’t work on mac and it DEFINITELY WILL NOT WORK ON WINDOWS). Make sure to read the wiki for installation and usage instructions. peace", "repo": "https://github.com/Chiron8/centaur", "demo": "https://github.com/Chiron8/centaur/wiki/2.-Install"}]...
|
[{"post_id": 85910, "ts": "202 [{"post_id": 85910, "ts": "2026-10-31T21:33:10Z", "hours": 1.3656, "text": "Just made a couple more bug fixes with the sandbox environment and staging directory management ahead of my final ship for stardance :). I also added tmux and its dependency, libevent."}, {"post_id": 85467, "ts": "2026-10-31T11:32:44Z", "hours": 0.9786, "text": "Quick devlog to mention that I updated how the bubblewrap environment works to include extra directories including /var and that I swapped out overlayfs to just binding directories as it makes it easier and more reliable :)"}, {"post_id": 82609, "ts": "2026-10-25T12:29:51Z", "hours": 13.1214, "text": "During the last three months, I significantly improved Centaur’s package-management infrastructure. I streamlined package source handling by adding automatic tarball downloads, extraction, and checksum verification. I also redesigned installation around a bubblewrap sandbox, staging directories, and manifests, which replaces the insecure system() function and makes package installation and remova safer. Alongside these changes, I improved uninstall usability, added x86-64 binaries, maintained the beta release workflow, and shipped versions 1.1.1 through 1.3.1."}, {"post_id": 22953, "ts": "2026-07-05T16:59:24Z", "hours": 3.9061, "text": "Centaur v1.1 Release Notes In this release, I focused on improving the .centaur package format and fixing several usability issues from v1.0. What I changed 1. I redesigned the .centaur syntax into 3 clear sections I changed .centaur files so package data is now split into: [meta] [install] [uninstall] This replaces the old mixed format where metadata, install steps, and uninstall script generation were all blended together. 2. I made metadata explicit and easier to parse In v1.0, package metadata and dependency handling relied on ad-hoc placement (for example, dependency lines and markers like === ADD DEPENDENCIES ABOVE THIS LINE === ). In v1.1, I moved this into a structured metadata block, for example: name = \"writedoc\" version = \"2.3-p1\" dependencies = \"example\" This makes package definitions cleaner and far more consistent. 3. I separated install and uninstall logic I moved installation commands into [install] and uninstall commands into [uninstall] . Previously, uninstall behavior was created indirectly during install by writing uninstall commands into another file. Now I define uninstall behavior directly in the package script itself, which is simpler and easier to maintain. 4. I fixed parser/runtime snags from v1.0 I resolved several pain points in .centaur files, including: line length limit issues newline/formatting sensitivity (such as needing blank lines between commands) general script parsing/execution reliability issues Result With v1.1, I made .centaur scripts more structured, more readable, and more reliable. The new format is easier for me (and others) to write, maintain, and parse going forward."}, {"post_id": 2848, "ts": "2026-06-05T18:08:43Z", "hours": 0.2842, "text": "I’m currently working on improving how I go about parsing package metadata. Previously, package dependencies were declared at the top of the file, however this soon limited adding extra metadata such as “preservation” type which I will implement later. I will now be using [meta] tags with variables inside them to allow for more advanced configuration!"}]...
|
Title: centaur by @chiron_dev8 | Stardance
Devlog Title: centaur by @chiron_dev8 | Stardance
Devlogs:
I’m currently working on improving how I go about parsing package metadata. Previously, package dependencies were declared at the top of the file, however this soon limited adding extra metadata such as “preservation” type which I will implement later. I will now be using [meta] tags with variables inside them to allow for more advanced configuration!
Centaur v1.1 Release Notes In this release, I focused on improving the .centaur package format and fixing several usability issues from v1.0. What I changed 1. I redesigned the .centaur syntax into 3 clear sections I changed .centaur files so package data is now split into: [meta] [install] [uninstall] This replaces the old mixed format where metadata, install steps, and uninstall script generation were all blended together. 2. I made metadata explicit and easier to parse In v1.0, package metadata and dependency handling relied on ad-hoc placement (for example, dependency lines and markers like === ADD DEPENDENCIES ABOVE THIS LINE === ). In v1.1, I moved this into a structured metadata block, for example: name = "writedoc" version = "2.3-p1" dependencies = "example" This makes package definitions cleaner and far more consistent. 3. I separated install and uninstall logic I moved installation commands into [install] and uninstall commands into [uninstall] . Previously, uninstall behavior was created indirectly during install by writing uninstall commands into another file. Now I define uninstall behavior directly in the package script itself, which is simpler and easier to maintain. 4. I fixed parser/runtime snags from v1.0 I resolved several pain points in .centaur files, including: line length limit issues newline/formatting sensitivity (such as needing blank lines between commands) general script parsing/execution reliability issues Result With v1.1, I made .centaur scripts more structured, more readable, and more reliable. The new format is easier for me (and others) to write, maintain, and parse going forward.
During the last three months, I significantly improved Centaur’s package-management infrastructure. I streamlined package source handling by adding automatic tarball downloads, extraction, and checksum verification. I also redesigned installation around a bubblewrap sandbox, staging directories, and manifests, which replaces the insecure system() function and makes package installation and remova safer. Alongside these changes, I improved uninstall usability, added x86-64 binaries, maintained the beta release workflow, and shipped versions 1.1.1 through 1.3.1.
Quick devlog to mention that I updated how the bubblewrap environment works to include extra directories including /var and that I swapped out overlayfs to just binding directories as it makes it easier and more reliable :)
Just made a couple more bug fixes with the sandbox environment and staging directory management ahead of my final ship for stardance :). I also added tmux and its dependency, libevent.
Ships:
I have worked on the project for a while before Stardance started. Probably around 95% of my hours are unlogged. I have made a package manager for UNIX based systems (specifically for my custom Linux based operating system, ChironOS). It can install, uninstall and update packages, manage dependencies, sync install files to the latest versions and automatically remove all unused packages. My favourite part of the project was finally seeing all the functions working together. I will still be working on this for a long long time as there is a lot I can improve on. My next goal will be to make package scripts for the whole of Linux From Scratch and Beyond LFS as this will prove the reliability of my package manager. If you want to test this please make sure you are using Linux (It will probably work on BSD but I haven’t tested it. It probably won’t work on mac and it DEFINITELY WILL NOT WORK ON WINDOWS). Make sure to read the wiki for installation and usage instructions. peace
Intro This is my second ship for centaur , my own source-based package manager. It can install, uninstall, manage dependencies, update, sync and clean packages. Please read the changelog below for all the awesome changes i’ve made! Attention reviewers Hi, thank you for taking the time to review my code, it means a lot. This package manager is only meant for linux operating systems which means that THIS WILL NOT WORK ON WINDOWS. If you do not want to set up a linux virtual environment to test this, please skip this project. I had a couple people complain that it couldn’t run on Windows and rated my project really poorly which isn’t very fair on me. Since this project deals with system tools it can potentially corrupt packages that are already installed on your system. Please consider using a virtual environment to test this. If you do not want to set up a linux virtual environment, and you are already running linux, please condider only installing writedoc using sudo centaur install writedoc as this is a package that I made myself so won’t corrupt anything you already have on your system. Changelog Quite a lot has happened since my last ship and I have taken onboard a lot of feedback from my reviewers (thank you!): Tarballs now do not manually have to be downloaded in the install section of the package script and all the necessary data such as the source link and sha256 checksum are included in the metadata. Tarballs are automatically checked for validity by comparing it’s known official checksum and the checksum of the downloaded file. Instead of blindly running commands directly on the host system, the commands are instead run through a secure sandbox environment using bubblewrap . This allows commands on the host computer to be used to configure and compile packages but files and directories that are not needed are read-only and cannot be modified by the packages. For example, to install htop , the make and bash commands can still be run but files and directories not related to the installation of it are read-only and cannot be modified or deleted. When a package tried to add new files to the host system, the sandbox intercepts them and adds it to a staging directory instead which can then be used for different purposes after installation. After installation is finished, the new files are then copied to the host system. A manifest file is now used to track what files were added to the host system when the package was installed. The staging directory is scanned and the files that are new are added to the manifest. When a user wants to uninstall a package, all the files that are referenced in the manifest are deleted. This means that the package script creators no longer need to make a separate uninstall block. A couple new packages have been added such as htop , tmux and libevent . There are loads more changes that are slightly less interesting but have improved the performance and readability of the code :)...
|
1793915422.4790187
|
1794001822.4790187
|
|
28734
|
cena by @harciufff | Stardance
|
harciufff
|
1
|
6.0
|
0
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-07-02T16:28:11Z", "hours": 6.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "What I Made I built “Cena”, an event-driven, edge-native system for real-time environment understanding. It runs a multi-threaded architecture that combines geometric telemetry (YOLO-World + MediaPipe) with asynchronous semantic descriptions (Moondream-2), logging everything locally into Markdown files inside Obsidian without vector database latency. The Challenge Managing heavy AI models locally on an edge device without performance throttling or memory leaks. I had to design a loosely-coupled asynchronous pipeline where the semantic layer runs on a background thread every 10 seconds, ensuring the geometric layer can maintain a smooth ~30 FPS with GPU Metal/MPS acceleration on my Mac M4. What I’m Proud Of The zero-indexing latency architecture. Instead of relying on heavy cloud vector databases, integrating the output directly into Obsidian markdown files creates a lightning-fast, local-first memory pipeline inspired by the LLM-as-an-OS concept. How to Test It Clone the GitHub repository, set up a Python virtual environment, and install the requirements. You will need a local camera feed and an Obsidian vault path configured in your script. Run python3 main.py to start the pipeline and press ‘q’ to safely shut down.", "repo": "https://github.com/harciufff/cena", "demo": "https://cenadev.vercel.app/"}]...
|
[{"post_id": 20862, "ts": "202 [{"post_id": 20862, "ts": "2026-07-02T16:09:06Z", "hours": 5.9483, "text": "Hey everyone! I’m building “Cena”, a local-first visual copilot system that provides real-time geometric telemetry and asynchronous semantic understanding. Note for the Shipwrights: I actually spent around 35-40 hours programming the core architecture and multi-threading structure last week before finding out about Stardance and installing the Hackatime extension. You can verify my daily development dates and progress through my GitHub commit history! How the system works: Geometric Layer (~30 FPS): Uses OpenCV to capture frames, feeding them to YOLO-World and MediaPipe Hands (running with GPU Metal/MPS acceleration on my Mac M4). Semantic Layer: A background thread invokes Moondream-2 every 10 seconds to get a text description of the scene. Agent Layer: As shown in the screenshot, the engine merges everything into plain markdown files inside Obsidian (LiveContext.md and PersistentContext.md). Inspired by Karpathy’s LLM-as-an-OS concept, this avoids vector database indexing latency on edge devices. The code is currently split into two versions (“Cena” and “CenaAmd”) representing the performance updates and VLM optimization of the pipeline. Check out the site here: https://cenadev.vercel.app/ Check out the full repository here: https://github.com/harciufff/cena"}]...
|
Title: cena by @harciufff | Stardance
Devlogs:
He Title: cena by @harciufff | Stardance
Devlogs:
Hey everyone! I’m building “Cena”, a local-first visual copilot system that provides real-time geometric telemetry and asynchronous semantic understanding. Note for the Shipwrights: I actually spent around 35-40 hours programming the core architecture and multi-threading structure last week before finding out about Stardance and installing the Hackatime extension. You can verify my daily development dates and progress through my GitHub commit history! How the system works: Geometric Layer (~30 FPS): Uses OpenCV to capture frames, feeding them to YOLO-World and MediaPipe Hands (running with GPU Metal/MPS acceleration on my Mac M4). Semantic Layer: A background thread invokes Moondream-2 every 10 seconds to get a text description of the scene. Agent Layer: As shown in the screenshot, the engine merges everything into plain markdown files inside Obsidian (LiveContext.md and PersistentContext.md). Inspired by Karpathy’s LLM-as-an-OS concept, this avoids vector database indexing latency on edge devices. The code is currently split into two versions (“Cena” and “CenaAmd”) representing the performance updates and VLM optimization of the pipeline. Check out the site here: https://cenadev.vercel.app/ Check out the full repository here: https://github.com/harciufff/cena
Ships:
What I Made I built “Cena”, an event-driven, edge-native system for real-time environment understanding. It runs a multi-threaded architecture that combines geometric telemetry (YOLO-World + MediaPipe) with asynchronous semantic descriptions (Moondream-2), logging everything locally into Markdown files inside Obsidian without vector database latency. The Challenge Managing heavy AI models locally on an edge device without performance throttling or memory leaks. I had to design a loosely-coupled asynchronous pipeline where the semantic layer runs on a background thread every 10 seconds, ensuring the geometric layer can maintain a smooth ~30 FPS with GPU Metal/MPS acceleration on my Mac M4. What I’m Proud Of The zero-indexing latency architecture. Instead of relying on heavy cloud vector databases, integrating the output directly into Obsidian markdown files creates a lightning-fast, local-first memory pipeline inspired by the LLM-as-an-OS concept. How to Test It Clone the GitHub repository, set up a Python virtual environment, and install the requirements. You will need a local camera feed and an Obsidian vault path configured in your script. Run python3 main.py to start the pipeline and press ‘q’ to safely shut down....
|
1793768167.2658184
|
1794632167.2658184
|
|
39727
|
celestenetv2 by @arborium | Stardance
|
arborium
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: celestenetv2 by @arborium | Stardance
|
1793450981.303177
|
1794055781.303177
|
|
52519
|
cehcklist-things organizer by @hereei | Stardance
|
hereei
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: cehcklist-things organizer by @hereei | Sta Title: cehcklist-things organizer by @hereei | Stardance...
|
1793444058.085248
|
1794048858.085248
|
|
42145
|
cdn™ by @e | Stardance
|
e
|
0
|
0.39
|
0
|
[]
|
[{"post_id": 37942, "ts": "202 [{"post_id": 37942, "ts": "2026-07-30T23:41:12Z", "hours": 0.3911, "text": "devlog 1: yay it works, the initial cdn handling code is done, so you can now actually upload and get files. currently there’s no actual way to upload it through a ui (ui is coming next probably) so you have to use curl but it works"}]...
|
Title: cdn™ by @e | Stardance
Devlogs:
devlog 1: Title: cdn™ by @e | Stardance
Devlogs:
devlog 1: yay it works, the initial cdn handling code is done, so you can now actually upload and get files. currently there’s no actual way to upload it through a ui (ui is coming next probably) so you have to use curl but it works...
|
1793821753.865404
|
1794426553.865404
|
|
27851
|
cdn™ by @e | Stardance
|
e
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: cdn™ by @e | Stardance
|
1793864931.4536648
|
1794728931.4536648
|
|
1616
|
cdn by @co | Stardance
|
co
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: cdn by @co | Stardance
|
1793390021.831892
|
1793994821.831892
|
|
22045
|
cctf.rs by @riverdalecoder | Stardance
|
riverdalecoder
|
1
|
36.28
|
391
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-07-09T18:05:43Z", "hours": 36.0, "multiplier": 11.29, "payout": 391, "blessing": "neutral", "text": "custom framework for a ctf platform written in rust. second ship on it will be the finalized platform instead of just the framework!! hope you enjoy!!", "repo": "https://github.com/TogiFerretFerret/cctf.rs", "demo": "https://ctf.river.hackclub.app/docs"}]...
|
[{"post_id": 25587, "ts": "202 [{"post_id": 25587, "ts": "2026-07-09T22:44:36Z", "hours": 0.2769, "text": "currently doing jp tl"}, {"post_id": 25416, "ts": "2026-07-09T18:00:54Z", "hours": 3.9936, "text": "ladida ship in a box"}, {"post_id": 23758, "ts": "2026-07-06T21:47:52Z", "hours": 3.8386, "text": "Fixed the import typos (FileService comma, Multipart import) Finished main.rs file_service wiring (with the gdrive RCLONE_REMOTE env override) + all test AppState sites Your routes + API_ROUTES entries are in (with your // TODO: make configurable on the body limit — left as-is) Added openapi /api/v1/files paths (multipart upload → ChallengeFile, download → binary) + ctf-file-missing key + English fluent strings Refactored the upload handler’s loop → while let (clippy never_loop)"}, {"post_id": 21913, "ts": "2026-07-04T02:52:15Z", "hours": 0.8875, "text": "i am locked in"}, {"post_id": 21885, "ts": "2026-07-04T01:54:05Z", "hours": 0.5717, "text": "wahoo this did not take 30 minutes (i kid this is not my only change)"}, {"post_id": 21864, "ts": "2026-07-04T01:06:37Z", "hours": 0.6411, "text": "CONCISEIFY"}, {"post_id": 21738, "ts": "2026-07-03T20:34:34Z", "hours": 0.5481, "text": "my payout should be at LEAST this much lmao (im kidding, almost)"}, {"post_id": 21701, "ts": "2026-07-03T20:00:01Z", "hours": 0.7075, "text": "i gave up on trying to write commit messages"}, {"post_id": 21663, "ts": "2026-07-03T18:52:33Z", "hours": 1.6639, "text": "yes i am starting to write real commit messages"}, {"post_id": 20606, "ts": "2026-07-02T07:39:00Z", "hours": 1.0633, "text": "tipscript"}, {"post_id": 20585, "ts": "2026-07-02T06:24:44Z", "hours": 1.0033, "text": "openapi"}, {"post_id": 20556, "ts": "2026-07-02T05:17:08Z", "hours": 0.69, "text": "i passed 20 hours lmao"}, {"post_id": 20513, "ts": "2026-07-02T03:51:04Z", "hours": 0.63, "text": "integration tests"}, {"post_id": 20480, "ts": "2026-07-02T02:41:29Z", "hours": 0.6553, "text": "MAIIIIIIN"}, {"post_id": 20464, "ts": "2026-07-02T01:49:59Z", "hours": 0.6083, "text": "me when doctests+refactor"}, {"post_id": 20413, "ts": "2026-07-02T00:30:37Z", "hours": 0.5308, "text": "email worker done"}, {"post_id": 20370, "ts": "2026-07-01T23:12:08Z", "hours": 1.4994, "text": "smtpp"}, {"post_id": 18934, "ts": "2026-06-30T03:35:14Z", "hours": 0.4983, "text": "pem; todolo"}, {"post_id": 18868, "ts": "2026-06-30T02:06:27Z", "hours": 1.4889, "text": "smtp; also checkout https://www.youtube.com/watch?v=PVwUMu1OVOo its really good"}, {"post_id": 17367, "ts": "2026-06-28T03:46:24Z", "hours": 0.5025, "text": "wow"}, {"post_id": 16042, "ts": "2026-06-26T02:04:24Z", "hours": 1.4167, "text": "whee all tests pass. still working on backend and excited to continue working on backend."}, {"post_id": 15963, "ts": "2026-06-25T22:30:14Z", "hours": 0.2867, "text": "i hate freeze time"}, {"post_id": 15704, "ts": "2026-06-25T15:16:48Z", "hours": 0.3469, "text": "rhai bad"}, {"post_id": 15685, "ts": "2026-06-25T14:34:31Z", "hours": 0.71, "text": "everything works except for this??? sigh"}, {"post_id": 15155, "ts": "2026-06-24T16:56:56Z", "hours": 0.4889, "text": "tests still pass, this time with some fixes to the instancer"}, {"post_id": 15121, "ts": "2026-06-24T15:57:56Z", "hours": 0.3278, "text": "yaya all tests pass"}, {"post_id": 15076, "ts": "2026-06-24T14:31:22Z", "hours": 0.3611, "text": "improved instance reaper"}, {"post_id": 14766, "ts": "2026-06-24T00:00:02Z", "hours": 0.6339, "text": "WIP Kubernetes instancer"}, {"post_id": 14424, "ts": "2026-06-23T15:29:45Z", "hours": 2.0428, "text": "something something testcases dont pass"}, {"post_id": 13714, "ts": "2026-06-22T14:12:25Z", "hours": 0.9958, "text": "rust is throwing a lot of errors, i might be debugging for a while; implemented scoreboardservice and added backend to support accuracy-based scoreboards"}, {"post_id": 13475, "ts": "2026-06-22T02:09:33Z", "hours": 0.4306, "text": "IT. IS. DONE (this one file lmao)"}, {"post_id": 13469, "ts": "2026-06-22T01:42:18Z", "hours": 0.4508, "text": "😭 i hate oauth"}, {"post_id": 13460, "ts": "2026-06-22T01:13:00Z", "hours": 1.1869, "text": "i hate rust. period."}, {"post_id": 13407, "ts": "2026-06-21T23:13:42Z", "hours": 0.3619, "text": "giant wip repos file"}, {"post_id": 13398, "ts": "2026-06-21T22:48:13Z", "hours": 0.7606, "text": "more types"}, {"post_id": 13236, "ts": "2026-06-21T17:50:50Z", "hours": 0.8161, "text": "got down all the types for challenges done, which will take a while to implement. however, it’s setup for modularity in the future which is good."}, {"post_id": 12806, "ts": "2026-06-20T22:58:50Z", "hours": 1.9642, "text": "i spent two hours writing a localized jwt system; this was DEFINITELY worth my time :trembling_thumbs_up: all jokes aside, very excited to be working on this project"}]...
|
Title: cctf.rs by @riverdalecoder | Stardance
Dev Title: cctf.rs by @riverdalecoder | Stardance
Devlogs:
i spent two hours writing a localized jwt system; this was DEFINITELY worth my time :trembling_thumbs_up: all jokes aside, very excited to be working on this project
got down all the types for challenges done, which will take a while to implement. however, it’s setup for modularity in the future which is good.
more types
giant wip repos file
i hate rust. period.
😭 i hate oauth
IT. IS. DONE (this one file lmao)
rust is throwing a lot of errors, i might be debugging for a while; implemented scoreboardservice and added backend to support accuracy-based scoreboards
something something testcases dont pass
WIP Kubernetes instancer
improved instance reaper
yaya all tests pass
tests still pass, this time with some fixes to the instancer
everything works except for this??? sigh
rhai bad
i hate freeze time
whee all tests pass. still working on backend and excited to continue working on backend.
wow
smtp; also checkout https://www.youtube.com/watch?v=PVwUMu1OVOo its really good
pem; todolo
smtpp
email worker done
me when doctests+refactor
MAIIIIIIN
integration tests
i passed 20 hours lmao
openapi
tipscript
yes i am starting to write real commit messages
i gave up on trying to write commit messages
my payout should be at LEAST this much lmao (im kidding, almost)
CONCISEIFY
wahoo this did not take 30 minutes (i kid this is not my only change)
i am locked in
Fixed the import typos (FileService comma, Multipart import) Finished main.rs file_service wiring (with the gdrive RCLONE_REMOTE env override) + all test AppState sites Your routes + API_ROUTES entries are in (with your // TODO: make configurable on the body limit — left as-is) Added openapi /api/v1/files paths (multipart upload → ChallengeFile, download → binary) + ctf-file-missing key + English fluent strings Refactored the upload handler’s loop → while let (clippy never_loop)
ladida ship in a box
currently doing jp tl
Ships:
custom framework for a ctf platform written in rust. second ship on it will be the finalized platform instead of just the framework!! hope you enjoy!!...
|
1793847246.6878223
|
1793933646.6878223
|
|
53976
|
caviar by @kelp | Stardance
|
kelp
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: caviar by @kelp | Stardance
|
1793947662.350222
|
1794811662.350222
|
|
47509
|
causal-interp by @ranveergill | Stardance
|
ranveergill
|
0
|
5.82
|
0
|
[]
|
[{"post_id": 53532, "ts": "202 [{"post_id": 53532, "ts": "2026-08-24T03:44:48Z", "hours": 5.8208, "text": "Devlog 1: Ten phases in, and the honest answer is half yes This project set out to build a system that can discover and causally validate the mechanisms behind a neural network’s behavior, aimed eventually at models too capable for a human to check by hand. Since no answer key exists for that eventual target, the whole approach only means anything if it’s validated against models where the answer key does exist and is publicly known. So that’s what the first ten phases did. I ran activation patching and path patching against GPT-2 small’s IOI circuit, a published, head-by-head known structure, and compared the results against the paper directly. Phase 1 recovered 20 of 26 heads. Phase 2 tested whether path patching would fix the misses. It didn’t, and I reported that as a negative result rather than reframing it. Phase 3 fixed 2 more heads using a threshold that was locked in before seeing any scores, specifically to avoid the trap of picking a number because it looks good after the fact. Phase 4 was the real turning point. Every phase up to that point told the system exactly where to look, which position, which part of the head, because the paper says so. Phase 4 made it search for that instead, with zero access to the answer key, and it converged on the right positions anyway. That’s the first real evidence this can do more than rediscover what it’s told. Phases 5 through 7 tested how far that holds up. The pipeline transferred to a second published circuit (greater than comparisons) with almost no code changes, and then to a third circuit on a completely different, smaller model architecture. Recovery dropped hard on that third one, from about 69 percent to 50 percent, and the cause turned out to be structural: the metric being used is blind to a whole class of heads by design, not by mistake. That’s a real limitation, not a bug, and it only became visible because the system was tested somewhere it hadn’t been tuned. Phases 8 through 10 tried to close that gap and the gaps around it. A system for flagging disagreement between different experimental setups caught the blind spot from phase 7 with no advance warning, which was genuinely exciting, but it also flagged plenty of things that turned out to be nothing, and nothing in the system could yet tell the two apart on its own. A separate attempt to have the system generate its own experiment setups from a one line description worked well at generating candidates, but the system still couldn’t reliably judge which candidate to trust without a human checking the answer. Writing the synthesis after all ten phases made the pattern obvious. Anything that comes down to measuring a number has been mechanized successfully. Anything that comes down to judging whether that number means something real still needs a person. That’s not a small gap. It’s arguably the one piece the entire project exists to eventually replace, since a model too capable to check by hand is exactly the situation where no one will be available to make that judgment call. A quick follow up check on old data confirmed this isn’t just a vague impression. It’s measurable, and it came back inconclusive on 13 data points, which is itself the honest answer at that sample size rather than something to round up into a finding. Next up is figuring out whether that judgment gap is solvable with a different kind of signal entirely, or whether it’s a genuine ceiling on this approach. Either answer would be useful to know for sure."}]...
|
Title: causal-interp by @ranveergill | Stardance
Title: causal-interp by @ranveergill | Stardance
Devlogs:
Devlog 1: Ten phases in, and the honest answer is half yes This project set out to build a system that can discover and causally validate the mechanisms behind a neural network’s behavior, aimed eventually at models too capable for a human to check by hand. Since no answer key exists for that eventual target, the whole approach only means anything if it’s validated against models where the answer key does exist and is publicly known. So that’s what the first ten phases did. I ran activation patching and path patching against GPT-2 small’s IOI circuit, a published, head-by-head known structure, and compared the results against the paper directly. Phase 1 recovered 20 of 26 heads. Phase 2 tested whether path patching would fix the misses. It didn’t, and I reported that as a negative result rather than reframing it. Phase 3 fixed 2 more heads using a threshold that was locked in before seeing any scores, specifically to avoid the trap of picking a number because it looks good after the fact. Phase 4 was the real turning point. Every phase up to that point told the system exactly where to look, which position, which part of the head, because the paper says so. Phase 4 made it search for that instead, with zero access to the answer key, and it converged on the right positions anyway. That’s the first real evidence this can do more than rediscover what it’s told. Phases 5 through 7 tested how far that holds up. The pipeline transferred to a second published circuit (greater than comparisons) with almost no code changes, and then to a third circuit on a completely different, smaller model architecture. Recovery dropped hard on that third one, from about 69 percent to 50 percent, and the cause turned out to be structural: the metric being used is blind to a whole class of heads by design, not by mistake. That’s a real limitation, not a bug, and it only became visible because the system was tested somewhere it hadn’t been tuned. Phases 8 through 10 tried to close that gap and the gaps around it. A system for flagging disagreement between different experimental setups caught the blind spot from phase 7 with no advance warning, which was genuinely exciting, but it also flagged plenty of things that turned out to be nothing, and nothing in the system could yet tell the two apart on its own. A separate attempt to have the system generate its own experiment setups from a one line description worked well at generating candidates, but the system still couldn’t reliably judge which candidate to trust without a human checking the answer. Writing the synthesis after all ten phases made the pattern obvious. Anything that comes down to measuring a number has been mechanized successfully. Anything that comes down to judging whether that number means something real still needs a person. That’s not a small gap. It’s arguably the one piece the entire project exists to eventually replace, since a model too capable to check by hand is exactly the situation where no one will be available to make that judgment call. A quick follow up check on old data confirmed this isn’t just a vague impression. It’s measurable, and it came back inconclusive on 13 data points, which is itself the honest answer at that sample size rather than something to round up into a finding. Next up is figuring out whether that judgment gap is solvable with a different kind of signal entirely, or whether it’s a genuine ceiling on this approach. Either answer would be useful to know for sure....
|
1793868481.7764602
|
1794732481.7764602
|
|
69868
|
catpad by @lollo22 | Stardance
|
lollo22
|
0
|
5.59
|
0
|
[]
|
[{"post_id": 81083, "ts": "202 [{"post_id": 81083, "ts": "2026-10-19T20:02:52Z", "hours": 5.5947, "text": "my catpad, an hackpad that resembles a cat i worked hard on this project, just need to make the firmware to complete the design"}]...
|
Title: catpad by @lollo22 | Stardance
Devlogs:
my Title: catpad by @lollo22 | Stardance
Devlogs:
my catpad, an hackpad that resembles a cat i worked hard on this project, just need to make the firmware to complete the design...
|
1793587639.4893079
|
1794192439.4893079
|
|
31699
|
catonpogo by @deg | Stardance
|
deg
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: catonpogo by @deg | Stardance
|
1793976772.783464
|
1794581572.783464
|
|
52440
|
catnipOS by @Pinidi | Stardance
|
Pinidi
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: catnipOS by @Pinidi | Stardance
|
1793440480.3243122
|
1794045280.3243122
|
|
27821
|
catching game by @emily_xm_hoang | Stardance
|
emily_xm_hoang
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: catching game by @emily_xm_hoang | Stardanc Title: catching game by @emily_xm_hoang | Stardance...
|
1793825705.3044639
|
1794430505.3044639
|
|
51571
|
catchat by @nafihvmuhammed | Stardance
|
nafihvmuhammed
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: catchat by @nafihvmuhammed | Stardance
|
1793397118.091067
|
1794001918.091067
|
|
71719
|
catch the star by @remychristiano | Stardance
|
remychristiano
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: catch the star by @remychristiano | Stardan Title: catch the star by @remychristiano | Stardance...
|
1793847293.7653744
|
1794452093.7653744
|
|
68182
|
catbord by @taco_taim | Stardance
|
taco_taim
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: catbord by @taco_taim | Stardance
|
1793738952.6053889
|
1794343752.6053889
|
|
17459
|
catOS by @xerneas | Stardance
|
|
1
|
8.0
|
0
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-07-18T02:29:48Z", "hours": 8.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "GatOS A personal operating system that runs in your browser and is, I will be honest, mostly about a cat. live: xerneas3318.github.io/GatOS code: github.com/xerneas3318/GatOS It is a little fake desktop. It has a top bar with a live clock, a background, and windows you can actually drag around, open, and close like a real OS. I built it from scratch over the last few weeks as my Stardance webOS project. What is on it welcome.exe : a short intro window (with a photo of my friend’s cat, obviously) dev.log : a notes-style reader with a sidebar of build entries, so you can read how the thing got made catnip : a full cookie-clicker game where the cookie is the cat. Tap the cat for treats, buy auto-petters and laser pointers and a “cat loaf mine” that earns treats while you are away. It saves your progress to your browser. Built with Plain HTML, CSS, and JavaScript. No framework, no build step. The whole color palette lives in one tiny tokens file thanks to CSS variables, so the entire theme changes from one place. Inter for the UI, Space Mono for the code and clock. What I learned Along the way I learned how to make an element draggable, how to keep windows stacking in the right order, how localStorage works, and how to split a project into a file structure a human can actually read instead of two giant cursed files. The cat, as always, remains. The cat is eternal. The cat now lives on the internet over HTTPS.", "repo": "https://github.com/xerneas3318/GatOS", "demo": "https://xerneas3318.github.io/GatOS/"}]...
|
[{"post_id": 29995, "ts": "202 [{"post_id": 29995, "ts": "2026-07-17T06:58:03Z", "hours": 4.4317, "text": "the files are files now opened this project after like a week away and genuinely could not find anything. one style.css , one script.js , both hundreds of lines of everything everywhere. i wrote every line of it and i still had to ctrl+f to find the clock. it was a junk drawer with a cat on top. so this whole session was cleaning. no new features, nothing shiny , just four hours of moving code around like a little file gremlin. what got split styles: one file for the top bar one for windows one for the desktop one per app same deal for the js, the clock, the dragging, the window stacking, the icons all live in their own files now. the two games moved into an apps/ folder so catnip stops rubbing up against the core window code (it’s a cat. it rubs on things.) the one smart thing the tokens file from last time made this way less painful than it deserved to be, since all the colors were already in one place, nothing broke visually while i shuffled everything. past me did one 1 smart thing and present me is grateful. weirdest part is opening a file and already knowing what’s in it before it loads. revolutionary technology.🤯 next time github pages: this thing needs a real url so it stops living exclusively on my laptop the cat remains. the cat is eternal. the cat’s code is just alphabetized now."}, {"post_id": 23853, "ts": "2026-07-07T01:21:34Z", "hours": 1.3867, "text": "NEW CAT MINI GAME first thing i did was get rid of the terminal look. it was cool for about a week, but staring at glowing teal on near-black was starting to feel like my OS was yelling at me. so i rebuilt the entire theme to look like a warm peach background, light grey panels, one orange accent doing all the heavy lifting. it’s calm now. it’s a website you could actually read at 2am without your retinas filing a complaint. did it the right way this time with CSS variables, so the whole palette lives in like eight lines at the top of the file. want a different accent? change one value. past me would have found-and-replaced forty hex codes and missed six of them. also swapped the font to Inter for that clean note-app feel, kept Space Mono around for code and the clock because some things are sacred. oh, and there’s a new app: catnip. it’s a cookie-clicker but the cookie is my friend’s cat. you tap the cat, treats go up, you buy auto-petters and laser pointers and eventually a “cat loaf mine” that prints treats while you’re not even looking. it saves your progress. i have lost dozens minutes of my life to it already and i wrote it. so the cat remains. the cat is eternal."}, {"post_id": 19491, "ts": "2026-06-30T20:24:20Z", "hours": 0.9378, "text": "the window moves now. you can grab welcome.exe by its title bar and fling it around the desktop, hit the little red dot to close it, and click xerneasOS up top to summon it back. it’s a real window now. it even has a title bar. spent a while wrestling with the drag math turns out you have to keep track of the cursor’s offset from the window when the drag starts. otherwise the window instantly teleports so its corner snaps to your mouse. not exactly the intended user experience. also did some housekeeping: ripped all the CSS into style.css and the JavaScript into script.js so my HTML isn’t one giant cursed wall of inline styles anymore. windows are now a reusable .window component, so making future apps should be way easier.oh, and there’s a live clock in the top bar that updates every second. new Date() my beloved. the cat remains. the cat is eternal. next time: a real wallpaper, maybe an actual second app, and slowly making this feel less like “one draggable window about a cat” and more like an operating system (that’s still about a cat)"}, {"post_id": 15979, "ts": "2026-06-25T23:01:38Z", "hours": 1.2875, "text": "today i learned the difference between margin and padding. profound stuff. life-changing, really. anyway none of that matters because i put a picture of my friend’s cat on my website and now it’s the best site ever made. just started webOS development! things i poked at: styled everything neon teal because i just can google fonts (VT323 my beloved) object-fit: cover so the cat doesn’t get crunched into a horror image border-radius so the cat has rounded corners now. very important. mission critical some might say next time: probably a wallpaper, a window, maybe a clock. or more cats. undecided."}]...
|
Title: catOS by @xerneas | Stardance
Devlogs:
tod Title: catOS by @xerneas | Stardance
Devlogs:
today i learned the difference between margin and padding. profound stuff. life-changing, really. anyway none of that matters because i put a picture of my friend’s cat on my website and now it’s the best site ever made. just started webOS development! things i poked at: styled everything neon teal because i just can google fonts (VT323 my beloved) object-fit: cover so the cat doesn’t get crunched into a horror image border-radius so the cat has rounded corners now. very important. mission critical some might say next time: probably a wallpaper, a window, maybe a clock. or more cats. undecided.
the window moves now. you can grab welcome.exe by its title bar and fling it around the desktop, hit the little red dot to close it, and click xerneasOS up top to summon it back. it’s a real window now. it even has a title bar. spent a while wrestling with the drag math turns out you have to keep track of the cursor’s offset from the window when the drag starts. otherwise the window instantly teleports so its corner snaps to your mouse. not exactly the intended user experience. also did some housekeeping: ripped all the CSS into style.css and the JavaScript into script.js so my HTML isn’t one giant cursed wall of inline styles anymore. windows are now a reusable .window component, so making future apps should be way easier.oh, and there’s a live clock in the top bar that updates every second. new Date() my beloved. the cat remains. the cat is eternal. next time: a real wallpaper, maybe an actual second app, and slowly making this feel less like “one draggable window about a cat” and more like an operating system (that’s still about a cat)
NEW CAT MINI GAME first thing i did was get rid of the terminal look. it was cool for about a week, but staring at glowing teal on near-black was starting to feel like my OS was yelling at me. so i rebuilt the entire theme to look like a warm peach background, light grey panels, one orange accent doing all the heavy lifting. it’s calm now. it’s a website you could actually read at 2am without your retinas filing a complaint. did it the right way this time with CSS variables, so the whole palette lives in like eight lines at the top of the file. want a different accent? change one value. past me would have found-and-replaced forty hex codes and missed six of them. also swapped the font to Inter for that clean note-app feel, kept Space Mono around for code and the clock because some things are sacred. oh, and there’s a new app: catnip. it’s a cookie-clicker but the cookie is my friend’s cat. you tap the cat, treats go up, you buy auto-petters and laser pointers and eventually a “cat loaf mine” that prints treats while you’re not even looking. it saves your progress. i have lost dozens minutes of my life to it already and i wrote it. so the cat remains. the cat is eternal.
the files are files now opened this project after like a week away and genuinely could not find anything. one style.css , one script.js , both hundreds of lines of everything everywhere. i wrote every line of it and i still had to ctrl+f to find the clock. it was a junk drawer with a cat on top. so this whole session was cleaning. no new features, nothing shiny , just four hours of moving code around like a little file gremlin. what got split styles: one file for the top bar one for windows one for the desktop one per app same deal for the js, the clock, the dragging, the window stacking, the icons all live in their own files now. the two games moved into an apps/ folder so catnip stops rubbing up against the core window code (it’s a cat. it rubs on things.) the one smart thing the tokens file from last time made this way less painful than it deserved to be, since all the colors were already in one place, nothing broke visually while i shuffled everything. past me did one 1 smart thing and present me is grateful. weirdest part is opening a file and already knowing what’s in it before it loads. revolutionary technology.🤯 next time github pages: this thing needs a real url so it stops living exclusively on my laptop the cat remains. the cat is eternal. the cat’s code is just alphabetized now.
Ships:
GatOS A personal operating system that runs in your browser and is, I will be honest, mostly about a cat. live: xerneas3318.github.io/GatOS code: github.com/xerneas3318/GatOS It is a little fake desktop. It has a top bar with a live clock, a background, and windows you can actually drag around, open, and close like a real OS. I built it from scratch over the last few weeks as my Stardance webOS project. What is on it welcome.exe : a short intro window (with a photo of my friend’s cat, obviously) dev.log : a notes-style reader with a sidebar of build entries, so you can read how the thing got made catnip : a full cookie-clicker game where the cookie is the cat. Tap the cat for treats, buy auto-petters and laser pointers and a “cat loaf mine” that earns treats while you are away. It saves your progress to your browser. Built with Plain HTML, CSS, and JavaScript. No framework, no build step. The whole color palette lives in one tiny tokens file thanks to CSS variables, so the entire theme changes from one place. Inter for the UI, Space Mono for the code and clock. What I learned Along the way I learned how to make an element draggable, how to keep windows stacking in the right order, how localStorage works, and how to split a project into a file structure a human can actually read instead of two giant cursed files. The cat, as always, remains. The cat is eternal. The cat now lives on the internet over HTTPS....
|
1793774850.3882294
|
1794638850.3882294
|
|
55462
|
cat-dle by @Camc | Stardance
|
Camc
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: cat-dle by @Camc | Stardance
|
1793821822.5991604
|
1794426622.5991604
|
|
38279
|
cat webOS by @connie_19 | Stardance
|
connie_19
|
1
|
2.0
|
0
|
[{"ship_num": 1, "ts": "2026-0 [{"ship_num": 1, "ts": "2026-07-27T12:59:06Z", "hours": 2.0, "multiplier": null, "payout": 0, "blessing": "neutral", "text": "cat themed webOS :)", "repo": "https://github.com/conniepocky/cat-webOS", "demo": "https://cat-webos.vercel.app/"}]...
|
[{"post_id": 35773, "ts": "202 [{"post_id": 35773, "ts": "2026-07-27T12:38:05Z", "hours": 0.2981, "text": "binary decoder! found a cute app icon of a cat with binary so decided i had to implement something for this. rather than just have the user type in binary numbers, i made it using a checkbox system so the user can understand how the number is being translated."}, {"post_id": 35753, "ts": "2026-07-27T12:08:18Z", "hours": 0.9436, "text": "made first two apps! notepad app - very simple allows to edit and saves data when open/close the app cat photo app - found an API that gives random cat photos from a database :) next step need to work on the rendering of these photos since some are very large and mess the window up a bit…"}, {"post_id": 32931, "ts": "2026-07-22T12:29:57Z", "hours": 0.7469, "text": "working on cat webOS submission :) very simple so far but have a few ideas for what to add next 😈"}]...
|
Title: cat webOS by @connie_19 | Stardance
Devlog Title: cat webOS by @connie_19 | Stardance
Devlogs:
working on cat webOS submission :) very simple so far but have a few ideas for what to add next 😈
made first two apps! notepad app - very simple allows to edit and saves data when open/close the app cat photo app - found an API that gives random cat photos from a database :) next step need to work on the rendering of these photos since some are very large and mess the window up a bit…
binary decoder! found a cute app icon of a cat with binary so decided i had to implement something for this. rather than just have the user type in binary numbers, i made it using a checkbox system so the user can understand how the number is being translated.
Ships:
cat themed webOS :)...
|
1793915765.861206
|
1794002165.861206
|
|
57415
|
cat pogo by @bako | Stardance
|
bako
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: cat pogo by @bako | Stardance
|
1793440055.5099363
|
1794044855.5099363
|
|
29706
|
cassette player music controller by @diomedea | St cassette player music controller by @diomedea | Stardance...
|
diomedea
|
0
|
2.45
|
0
|
[]
|
[{"post_id": 29978, "ts": "202 [{"post_id": 29978, "ts": "2026-07-17T06:10:08Z", "hours": 1.1669, "text": "remade the schematic and pcb!! i figured if i was adding so much to it and changing the layout so much i might as well just make a new one. so i did"}, {"post_id": 28747, "ts": "2026-07-15T06:35:16Z", "hours": 0.7833, "text": "got really busy for a couple weeks with school stuff but i Finally got back to this! i realised i was just making a Normal Tiny Keyboard. and thats BORING. so i started thinking what i might actually use something like this for and came up with this!! it’s just gonna have macros for playing pausing and skipping music thru global shortcuts (with powertoys or something) that is all devlog finished"}, {"post_id": 20626, "ts": "2026-07-02T08:31:47Z", "hours": 0.5, "text": "i am Working on Keyboard! ive never used fusion360 before i am fighting for my life out here if anyone knows how to resize the keyboard plate from that website generator PLEASE tell me im losing my mind"}]...
|
Title: cassette player music controller by @diomed Title: cassette player music controller by @diomedea | Stardance
Devlogs:
i am Working on Keyboard! ive never used fusion360 before i am fighting for my life out here if anyone knows how to resize the keyboard plate from that website generator PLEASE tell me im losing my mind
got really busy for a couple weeks with school stuff but i Finally got back to this! i realised i was just making a Normal Tiny Keyboard. and thats BORING. so i started thinking what i might actually use something like this for and came up with this!! it’s just gonna have macros for playing pausing and skipping music thru global shortcuts (with powertoys or something) that is all devlog finished
remade the schematic and pcb!! i figured if i was adding so much to it and changing the layout so much i might as well just make a new one. so i did...
|
1793418522.9114635
|
1794023322.9114635
|
|
25239
|
casio fx-83GT X drum machine by @Jamothdodgeth | S casio fx-83GT X drum machine by @Jamothdodgeth | Stardance...
|
Jamothdodgeth
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: casio fx-83GT X drum machine by @Jamothdodg Title: casio fx-83GT X drum machine by @Jamothdodgeth | Stardance...
|
1793879583.594551
|
1794484383.594551
|
|
4141
|
casino by @harshu_yelle | Stardance
|
harshu_yelle
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: casino by @harshu_yelle | Stardance
|
1793901122.7377872
|
1794505922.7377872
|
|
57373
|
cashly by @sidhhrana5 | Stardance
|
sidhhrana5
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: cashly by @sidhhrana5 | Stardance
|
1793436526.9774835
|
1794041326.9774835
|
|
64462
|
case intersellar robot by @AlexL0103 | Stardance
|
AlexL0103
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: case intersellar robot by @AlexL0103 | Star Title: case intersellar robot by @AlexL0103 | Stardance...
|
1793674420.1456974
|
1794279220.1456974
|
|
64110
|
carting game by @omaralalem105 | Stardance
|
omaralalem105
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: carting game by @omaralalem105 | Stardance
|
1793670449.1482732
|
1794275249.1482732
|
|
60693
|
carsim by @jensel | Stardance
|
jensel
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: carsim by @jensel | Stardance
|
1793645474.7855124
|
1794250274.7855124
|
|
49457
|
care nest by @sakethramt | Stardance
|
sakethramt
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: care nest by @sakethramt | Stardance
|
1793285607.8728175
|
1793890407.8728175
|
|
22273
|
cardTrader by @Fa2002 | Stardance
|
Fa2002
|
0
|
0.0
|
0
|
[]
|
[]
|
Title: cardTrader by @Fa2002 | Stardance
|
1793559274.3242679
|
1794164074.3242679
|
|
6580
|
card video game by @flyingfish | Stardance
|
flyingfish
|
0
|
32.22
|
0
|
[]
|
[{"post_id": 9627, "ts": "2026 [{"post_id": 9627, "ts": "2026-06-16T01:05:29Z", "hours": 8.2886, "text": "i added a bot that can play against you. it really is just a random number generator but it works. i also made the plus 2 cards and skip cards work along with fixing a bug where reverse cards cant be played on the sides of the board(unity takes a long time to load and i didnt get a new screen shot so i just put the old one in mb)"}, {"post_id": 6603, "ts": "2026-06-11T02:30:51Z", "hours": 9.5603, "text": "i fix the bug where it would crash if you place any card on the side and i made the reverse card work. reverse cards switch the card above and below and it also switches sides"}, {"post_id": 5376, "ts": "2026-06-09T03:49:52Z", "hours": 5.0986, "text": "i just spent 3 hours making it so you cant place a card on another card and i added some logic so you can only place cards next to cards of the same color or number"}, {"post_id": 4461, "ts": "2026-06-07T19:46:11Z", "hours": 9.2722, "text": "i forgot to post any dev logs but so far i have made a grid that you can put cards on but there is no logic so its mostly visual rn"}]...
|
Title: card video game by @flyingfish | Stardance
Title: card video game by @flyingfish | Stardance
Devlogs:
i forgot to post any dev logs but so far i have made a grid that you can put cards on but there is no logic so its mostly visual rn
i just spent 3 hours making it so you cant place a card on another card and i added some logic so you can only place cards next to cards of the same color or number
i fix the bug where it would crash if you place any card on the side and i made the reverse card work. reverse cards switch the card above and below and it also switches sides
i added a bot that can play against you. it really is just a random number generator but it works. i also made the plus 2 cards and skip cards work along with fixing a bug where reverse cards cant be played on the sides of the board(unity takes a long time to load and i didnt get a new screen shot so i just put the old one in mb)...
|
1793292854.09978
|
1793897654.09978
|