Whiteboard Concept Explainer
Un concetto, disegnato mentre viene detto.
Trasforma la tua AI in un produttore di video esplicativi: trasforma un concetto difficile in una breve sceneggiatura per video lavagna in cui una mano disegna ogni idea mentre la voce narrante la dice, con un disegno per beat, parole semplici e una struttura cancella-e-ricostruisci che tiene lo spettatore incollato.

Installala in 3 passaggi
Nessuna registrazione. Nessuno strumento da riga di comando. Scegli la tua AI, copia una riga e incollala.
Preferisci fare da te? Acquista la skill ($4.99), scarica lo ZIP dalla conferma d'acquisto, decomprimilo nella cartella mostrata sopra e riavvia la sessione della tua AI — stesso risultato. L'hai già acquistata? Il tuo assistente AI può scaricarla per te.
Cosa contiene
Un file SKILL.md (la guida operativa) + un modello di partenza. Un'anteprima:
Whiteboard Concept Explainer
When to use
- A concept is hard to follow in prose — a process, a trade-off, a system with moving parts — and a reader needs to watch it get drawn one piece at a time
- A team needs a short video (roughly 90 seconds to 3 minutes) that explains one idea to newcomers: how a product works, how a policy works, why a decision was made
- A lesson, onboarding video, or pitch needs a format that is cheap to produce, easy to revise, and watchable without sound design or footage
- A presenter keeps losing the audience in slide decks: the audience reads ahead, zones out, or cannot see how the pieces connect
- A concept has a "draw it on a napkin" version that everyone understands, and the video should be that napkin drawing, done properly
Workflow
- State the one concept and the one takeaway (input: the topic → output: a concept + takeaway sentence)
Pick one concept and write the takeaway as a single sentence the viewer should be able to repeat after watching: "Compound interest means your interest earns interest." If the takeaway needs an "and" to finish, it is two concepts — split it into two videos. Write down what the viewer already knows (the starting point) and the one confusion the video must clear (the misconception), because the script is built to move the viewer from the first to the second. A video that tries to explain everything about a topic explains nothing; a video that clears one confusion earns a rewatch. - Break the concept into drawable beats (input: concept + takeaway → output: a beat list of 4–7 drawings)
Split the concept into four to seven beats, each expressible as one simple drawing: a labeled box, an arrow, a two-column comparison, a loop, a before/after pair. One beat, one drawing — if a beat needs two drawings, it is two beats. Order the beats so each drawing builds on the previous one on the same board: draw the parts, then connect them, then label the whole. Write each beat as "drawing + one spoken sentence" (for example: "Draw two piles, one growing slowly, one curving upward. Say: 'The second pile grows on its own growth.'"). A beat list that cannot be drawn in simple shapes is a sign the concept is still too abstract — simplify the concept, not the drawing. - Write the voiceover to the drawings (input: beat list → output: a timed voiceover script)
Write the voiceover so the words and the drawing land together: the narrator says "this pile" exactly as the hand draws that pile, never three sentences ahead or behind. Use plain spoken words — short sentences, everyday nouns, no jargon without an immediate plain-language translation. Read the script aloud and time it: aim for roughly 130–150 spoken words per minute, and keep the total to the length the concept deserves (most concepts land in 90 seconds to 3 minutes; a script that runs past 4 minutes should lose a beat, not gain background music). Mark every erase or board-clear in the script, because an erase is a scene change the viewer must be ready for. - Plan the board, the hand, and the erases (input: voiceover script → output: a board plan)
Plan the board as a sequence, not a single picture: decide what is on the board at each beat, what stays, what gets erased, and what gets rebuilt. Keep the board uncluttered — erase a finished section before starting the next, and never leave more than three labeled elements on screen at once. Decide the hand's path (draw left to right, top to bottom, in the order the voiceover mentions things) and the drawing style (one marker weight for structure, one for labels, one accent color at most for the key insight). If the video uses a real hand or a digital hand, the rules are the same: the hand draws what the voice says, when the voice says it, and never covers the drawing it just made. - Record, review for sync, and ship with captions (input: board plan + script → output: a finished video + captions)
Record the voiceover first, then animate or film the drawings to match it — syncing drawings to a finished voice track is far easier than syncing a voice to finished drawings. Review the cut three times: once with sound on (do words and drawings land together), once with sound off (does the drawing sequence alone tell the story), once at double speed (does any beat drag or flash past). Fix sync slips before polishing anything else — a half-second mismatch is more confusing than a rough drawing. Ship with captions (accurate, timed, not auto-generated-and-unchecked) and a one-line description carrying the takeaway sentence, so the video works muted, searchable, and accessible.
Anteprima del modello
Whiteboard Concept Explainer — Script Template
One video = one concept, one takeaway. Fill this in before recording anything.
Concept: {name}
Takeaway sentence (viewer repeats after watching): {one sentence} Viewer already knows: {starting point} Misconception to clear: {the confusion the video must clear} Target length: {90 sec – 3 min} | Word budget: {~130–150 words per minute}
Beat list (4–7 beats, one drawing + one spoken sentence each)
Beat 1
Drawing: {one simple drawing — box, arrow, comparison, loop, before/after} Spoken sentence: "{the one sentence said as the hand draws}" Stays on board: {yes / no} | Erased after: {yes / no}
Il pacchetto della skill (SKILL.md + modello) è scritto in inglese.


