AnkiWeb
- Votes
- 0 (π 0 Β· π 0)
- Updated
- 2026-10-25
- Anki versions
- 2.1.1~
- Languages
- JavaScript, JSON, Python
- Description language
- en
AnkiWeb addon 2052537765
Turns Anki into a speedcubing trainer by generating cube case diagrams and animated algorithm playback from typed algorithms, with bundled CFOP sets and custom colour schemes.
Open on AnkiWeb GitHub Ask about alternatives
active
| Min Anki | Max Anki | Updated |
|---|---|---|
| 2.1.1 | + | 2026-10-25 |
Loadingβ¦
Creates interactive medical flowcharts and mind maps inside Anki, with drag-and-drop editing, auto-cloze cards, JSON/AI import, alignment tools, and click-to-reveal review.
Displays a randomly selected entry from a chosen field each time a card loads, using pipe-separated values via {{rand-alg:(field name)}}.
Creates interactive drag-and-drop flashcards where you label images or sort items into categories, with instant feedback, shuffling, theme and layout options, across Anki Desktop, AnkiDroid and AnkiMobile.
Imports chess PGN game collections into Anki decks, creating opening-repertoire flashcards with board diagrams, annotations, and nested variations, while allowing re-imports without losing learning progress.
Automatically adds colored stroke order diagrams to kanji card fields, storing them as media so they display on AnkiWeb even without the addon installed.
Gamifies Anki study by converting review streaks and card learning into gold, farm progress, and turn-based battles, including leaderboards and multiplayer matches.
CubeCards turns Anki into a speedcubing algorithm trainer. Type an algorithm once and the add-on draws the Rubik's Cube case and animates the solution. No screenshots, and the picture can never disagree with the alg.
Includes the full CFOP algorithm sets: F2L (41), OLL (57), PLL (21) and COLL (40), sorted into sub-decks by case group (for example OLL Dot Case, COLL Sune, PLL EPLL), plus 2-look OLL.
Name | algHow to use: Tools > Cube Trainer > Import algorithm sets. Change colours and views under Tools > Cube Trainer > Settings, then sync.
Early release: bug reports and suggestions welcome. Source code, screenshots and docs: [your GitHub link]
Flashcards for Rubik's Cube algorithms with cube pictures generated from the algorithm and an animated player. You type the algorithm once. The add-on works out the case, draws it, and animates the solution. The picture and the algorithm can never disagree, and you never have to screenshot a case again.
| Question side | Answer side (3D, animated) | Any colour scheme |
|---|---|---|
![]() |
![]() |
![]() |
oll, coll, cmll, f2l, cross, or full).OLL::Dot Case, COLL::Sune, PLL::EPLL).Name | algorithm lines to make cards for COLL, ZBLL, F2L, or anything else.From AnkiWeb: Tools > Add-ons > Get Add-ons..., enter the code shown on the add-on's AnkiWeb page, and restart Anki.
From a file: download cubetrainer.ankiaddon, then Tools > Add-ons > Install from file....
Phone only, or no add-on: import Cube_Trainer_starter.apkg. It carries the note type and all bundled cards, and needs no add-on to review.
Requires Anki 2.1.50 or newer. The add-on itself runs on desktop only. Anki add-ons cannot run on mobile, but the cards it creates sync to your phone like any other cards.
| Card | Question | Answer |
|---|---|---|
| Case to Alg | The case picture | Name, 3D cube, animated algorithm, notes |
| Alg to Case | The algorithm as text (only for notes whose Reverse field is filled) | The case picture |
| Field | Meaning |
|---|---|
| Name | Shown on the answer, for example OLL 21 or Sune 3 |
| Set | Small label on the question, for example OLL or PLL |
| Alg | The algorithm. The case is derived from it automatically |
| Scramble | Optional. Use this instead of "the inverse of the alg", for example for algs that only solve part of the cube |
| Stage | full, oll, coll, cmll, f2l or cross |
| View | Blank or top = top face, bottom = turn the cube over to show the D face, flip = 3D view with the cube flipped, 3d = 3D only (used by the F2L cards) |
| Cross | Optional per-note cross colour: white, yellow, green, blue, red or orange |
| Orient | Optional per-note extra whole-cube rotation for the picture, for example y or y2 |
| Notes | Free text shown on the answer (HTML allowed) |
| Reverse | Put any text here to also get an Alg to Case card |
Supported: face turns U D F B R L, wide moves (r or Rw), slices M E S, rotations x y z, primes (R'), double turns (R2, R2'), and parentheses (ignored).
Not supported: commutator notation such as [A, B]. Expand those to plain moves first.
A mistyped move shows a small red message on the card, for example Cube viewer: Unknown move: Q.
If an alg contains cube rotations (such as the leading x in some PLL algs), the case is drawn with the last layer on top, which is what you want.
oll shows only which stickers are oriented. coll shows the corners in full and the edges as oriented or not. cmll shows corners only. f2l hides the last layer. cross shows only the cross pieces. full shows everything.| Set | Cases | Sub-decks |
|---|---|---|
| F2L | 41 | Free Pairs, Disconnected Pairs, Connected Pairs, Corner In Slot, Edge In Slot, Pieces In Slot |
| OLL | 57 | Dot Case, Square Shapes, Lightning Shapes, Fish Shapes, Knight Move Shapes, OCLL, All Corners Oriented, Awkward Shapes, P Shapes, T Shapes, C Shapes, W Shapes, L Shapes, Line Shapes |
| PLL | 21 | Adjacent Swap, Opposite Swap, EPLL |
| COLL | 40 | Sune, Anti-Sune, H, L, Pi, T, U |
| 2-look OLL | 10 | Edges (3) and corners (7) |
f2l stage, which greys out the last layer.tests/test_sets.js against an independent cube model: F2L pieces stay intact, OLL has 57 distinct patterns, PLL has 21 distinct cases, COLL algs solve the corners with edges oriented, and every F2L case is a single-slot case.The algorithm choices follow the lists on SpeedCubeDB by Gil Zussman. Individual algorithms are move sequences shared across the cubing community, but if you redistribute this add-on, check SpeedCubeDB's terms and credit them. The add-on's code is MIT licensed.
Open Tools > Cube Trainer > Settings... for the common options (cross colour, front colour, views, speed, autoplay, loop). Everything else is in Tools > Add-ons > Cube Trainer > Config:
| Option | Default | Meaning |
|---|---|---|
cross_color |
white |
Colour on the face-down (D) side: white, yellow, green, blue, red, orange |
front_color |
"" |
Colour on the front face, or automatic |
colors |
standard | Hex value for each colour name. Match your own stickers, for example "red": "#d11b79" |
grey_color |
#8a8a8a |
Colour of pieces greyed out by a stage |
body_color |
#161616 |
Cube body and the gaps between stickers |
view |
top |
Default picture orientation: top, bottom or flip |
stage |
"" |
Default stage for notes with an empty Stage field |
orient |
"" |
Extra whole-cube rotation for every picture, for example y |
front_show / back_show |
flat / both |
What the question / answer side shows: flat, 3d or both |
speed, move_ms |
1, 420 |
Animation speed multiplier and quarter-turn length in milliseconds |
autoplay, loop |
true, false |
Animation behaviour on the answer side |
reverse_cards |
false |
Also create Alg to Case cards when importing |
deck_root |
Cube Trainer |
Parent deck used by the import dialogs |
default_sets |
2-look OLL | Sets pre-ticked in the import dialog |
Settings are baked into the card templates when you press Save (or run Install / update note type), so sync afterwards to update your phone. Your notes are never changed.
Anki add-ons cannot run on mobile, so the add-on only creates the note type and notes. The drawing and animation code (plain JavaScript and SVG, about 30 KB) is stored inside the card templates, so it works wherever your collection syncs. There is no WebGL, no media files and no network access.
Anki treats {{ and }} in templates as field tags, so the script is sanitised before it is inlined.
No network access, no telemetry, no data leaves your computer. The add-on only reads and writes your Anki collection (the "Cube Case" note type and the notes you import) and its own config.
pip install -r requirements-dev.txt
npm install
bash tests/run_all.sh # rebuilds the sets, then engine, algorithm, Anki render, browser (jsdom) and GUI tests
python3 build.py # writes dist/cubetrainer.ankiaddon and the starter deck
Layout: cubetrainer/ is the add-on (core.py has all the logic and no GUI imports, __init__.py is the menu and dialogs, cubeviz.js is the cube engine, renderers and player). tools/cfop_algs.txt is the algorithm list and tools/build_sets.js turns it into cubetrainer/algsets.json. See CHANGELOG.md for history.
Issues and pull requests are welcome, especially: more algorithm sets, notation support, and reports from different Anki versions and devices.
MIT, see LICENSE.