โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
Teach AI How to Act
You teach it how to do โ and it tries.
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
[/p]A little desktop toy where you teach an AI how to do something โ just by typing.
You type out a thing โ how to take a shower, which hand goes in first โ and the AI takes a guess, tries to figure out what you meant, and shows you what it came up with. Whether it got it right is up to you to judge.
It doesn't actually learn anything. It just tries to understand what you wrote, and then it shows you how much it understood. You decide if it did it right.
๐ฌ What's the deal
๐ You say one line ยท It takes one shot
You type "first reach with the left hand" / "now raise the right" โ the AI reads the line, flips through whatever it knows, and picks something to play back. You watch and decide: did that match what I meant? If not, keep teaching.}
๐๏ธ You teach โ type, send, tell it what to do
๐ง It tries โ reads your words, digs through its own library, picks something
๐๏ธ You watch โ it shows you what it found, you judge if it's right
๐ฅ How to see what it picked
Full-screen video is how it shows you the result โ no menus, no controls, just the clip it chose. You watch and see if it matches what you tried to teach. The chat drawer sits quietly in the corner until you need it.}
๐๏ธ Seamless cuts โ segments preload so there's no blackout or hiccup
๐ฒ Doesn't block โ the teach drawer stays out of the way until you expand it
๐ฑ๏ธ Click and teach โ pick a series from the home page, start teaching right away
๐ค Local model ยท fully offline
Everything runs on your own machine. No internet needed, no account, no uploads. The local LLM (a GGUF model you can replace) does the guessing right here.}
โก Plug and play โ first launch wakes the local model automatically
๐ Swappable โ drop any GGUF model you like into the models folder
๐ก๏ธ Privacy โ every teach goes nowhere but your machine
๐ฅ๏ธ Minimal interface
Two states: pick a series โ watch full-screen + type in the teach drawer. That's it. Nothing else gets in the way.}
๐ฑ๏ธ Click a card to enter a series, click the teach drawer to type
โจ๏ธ F11 fullscreen / Esc back to home / Enter to open the teach drawer
๐ฎ Legacy W/D/S/A frame controls are still there if you need frame-by-frame inspection
๐ง Tech stack glance
โผ 01๏ฝUser layer
Series picker ยท Full-screen player ยท Teach drawer ยท Queue HUD ยท Context menu
โผ 02๏ฝAI teach core
AIDirector ยท Local LLM ยท Instruction parse ยท Action match ยท Fallback chain
โผ 03๏ฝShow engine
Dual-video pre-swap ยท No-gap transition ยท Buffer preload ยท Event sync
โผ 04๏ฝState layer
Play queue ยท Current index ยท History cache ยท Two-way sync
โผ 05๏ฝLocal AI runtime
llama-server ยท GGUF model ยท Port 9141 ยท Auto start/cleanup
โผ 06๏ฝData layer
Local series config ยท Local playlist ยท Video assets ยท Fully offline
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
This software includes modified Tauri 2 application, llama.cpp (MIT) by ggml-org, Hunyuan 7B by Tencent (see model card for license), and FFmpeg (LGPL) as external tool. Full license texts are included in the installation directory.
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
You teach it how to do something.
It tries to understand you.
Then it shows you what it got.
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
