thinkhard.SI

Prompt repository Image & video

Animated scene with a character for an explainer video

Your explainer video needs a short animated scene of a character in a relatable situation, in a style that stays the same from scene to scene.

Registry no.
MED-050
Collection
Image & video
Laid out for
Video model
Score
10 out of 10
Blanks to fill
11

Image & videoVideo model

The prompt

Score 10 of 10, Excellent

Use it when Your explainer video needs a short animated scene of a character in a relatable situation, in a style that stays the same from scene to scene.

  • {{topic}}
  • {{use_case}}
  • {{scene_purpose}}
  • {{narration}}
  • {{character}}
  • {{action}}
  • {{setting}}
  • {{palette}}
  • {{duration_seconds}}
  • {{aspect_ratio}}
  • {{sound_effects}}
Full prompt and why it works
Prompt
A short 2D animated scene for an explainer video about {{topic}}, for {{use_case}}. The scene shows {{scene_purpose}}, so the viewer recognizes the situation in a few seconds, before the narration explains it.

The narration that plays over this scene, for timing and mood only:
"""
{{narration}}
"""

Character: {{character}}. If I attach an image of this character or a frame from an earlier scene, it is the reference: keep the character's design, proportions, clothing and colors exactly as shown.

Action: {{action}}. One clear action with a readable change in the character's expression or posture, acted broadly enough to read without words.

Setting: {{setting}}, kept simple, with only the objects the action needs.

Camera: a static medium shot, or a single slow push-in if the moment needs emphasis, with no cuts.

Style: flat 2D vector animation with soft rounded shapes, smooth easing on every movement, gentle flat lighting with soft shadows, and a limited palette of {{palette}}, in the style of a friendly modern explainer video. The scene should feel warm and easy to follow, and the character stays on model from the first frame to the last.

Timing and format: {{duration_seconds}} seconds, one continuous shot, {{aspect_ratio}}.

Sound: {{sound_effects}}, and soft instrumental music low in the mix. No voice-over and no speech, since narration is added in editing.

Leave out all text, letters, numbers, speech bubbles, interface text, logos and watermarks, because generated lettering comes out garbled; add any words in your editor. Do not make the character resemble any existing cartoon or brand character. If the action is unclear or would need more than one shot, show only its first moment, slowly and simply, rather than adding new characters or events.

Variables

{{topic}}
What the video explains, for example "how our app splits shared bills".
{{use_case}}
Where it will be shown, for example "our home page" or "a sales presentation".
{{scene_purpose}}
What the scene shows, for example "the problem: someone frustrated while working out who owes what after a group dinner".
{{narration}}
The narration line that plays over this scene, so the action can match it, or "none".
{{character}}
A fictional character with fixed traits, for example "a young man with curly black hair, a green hoodie and round glasses".
{{action}}
What the character does, for example "stares at a long receipt, then rubs his forehead and sighs".
{{setting}}
Where it happens, for example "a table in a cozy restaurant after dinner".
{{palette}}
Two to four colors, for example "navy and sky blue with a warm orange accent".
{{duration_seconds}}
A clip length your video model supports; check its settings.
{{aspect_ratio}}
16:9 for a web page or slide, 9:16 for vertical social video.
{{sound_effects}}
One or two sounds, for example "the rustle of paper and a soft sigh".

Why it works

  • An example. An attached image of the character, or a frame from an earlier scene, is what keeps the design the same across the scenes of one explainer.
  • Background facts. The narration line is given for timing and mood only, so the action matches what the viewer will hear while no speech is generated.
  • What to avoid. No text or speech bubbles in the frame, since video models render lettering poorly, and no resemblance to existing characters.

How to use it

  1. Press Fill in blanks and type your details into each blank. What you type stays in this browser.
  2. Copy the filled prompt into the AI tool you already use, such as ChatGPT, Claude or Gemini, or open it in the Grader or the Prompt Tester here.
  3. Before you rely on it, try it on three or four real examples. The Prompt Tester runs them all at once and shows which ones go wrong.
Use it from code

Get this prompt as JSON from the free REST API, with your own text in the blanks, or let an AI assistant fetch it through the MCP server. No key.

curl -s https://thinkhard.si/api/v1/prompts/explainer-animation-character-scene

With the MCP server added to Claude, ChatGPT, Cursor or VS Code, call get_prompt with {"prompt": "explainer-animation-character-scene"}, or just ask for it by name.