- Make it Pop by Khulan
- Posts
- Make it Pop #07 - Steal my "Get Lit" Christmas jumper prompts 🎄
Make it Pop #07 - Steal my "Get Lit" Christmas jumper prompts 🎄
The full prompt list from my 'Unwrap Your Creativity' session on Google Cloud Tech Youtube channel
Hello my favorite people! đź‘‹
I had so much fun joining Stephanie Wong live this week for our 'Unwrap your creativity: A gen media special' on Google Cloud Tech Youtube channel. We took a simple idea, a 'Get Lit' Christmas jumper, and turned it into a full digital brand. Below, I’m sharing the full breakdown, including prompts and outputs we got, so you can see exactly how the gen media stack comes together.
We used Gemini 3, Nano Banana Pro, Veo, Gemini TTS and Lyria to build brand guidelines, global ad campaign, the video, and vibe coded a website. We are starting from 1 jumper image. I was using Vertex AI platform for this demo.
Reminder: I rarely write my own prompts. I talk to Gemini 3, literally yap away (using voice function) giving it all the creative direction I am envisioning, what I want, what I don’t want…. Then I have Gemini 3 organize my yapping and help me create text prompts for my images, videos, audio and voice overs.
Our reference source image:
Step 1: Brand Guideline (Gemini 3 + Nano Banana Pro)
I uploaded the photo of the jumper into Gemini 3 and asked it to act as my Creative Director to establish the brand identity first.
Image reference + prompt
 | “You are an expert Creative Director. I am launching a new fashion brand based on this Christmas jumper. Based on the visual style of this image, please generate a comprehensive text-based Brand Guideline. Include a Brand Name, a specific Color Palette (with hex codes), and a Tone of Voice description." |
Output:
You get a long descriptive text output on the brand guideline (too long to copy paste here in this newsletter). I get best text fidelity Nano Banana Pro results when I give it an input like PDF, hence here I am starting first with Gemini 3 since it’s multimodal and can analyze an image (also video and audio btw).
Then switch to Nano Banana Pro and paste this prompt:
Prompt:
“Now, create a visual representation of the brand guideline you created for me”
Output:
Step 2: The Visuals (Nano Banana Pro)
We now have a "visual source of truth." I fed those guidelines into Nano Banana Pro to create a consistent look for everything else we generate. Now we can go global with an ad campaign.
2 reference images + prompt
 Jumper |  Brand guideline | "Create a cute, 3D isometric knitted wool character, a friendly badger wearing the 'Get Lit' jumper, standing in a snowy, knitted wool London street scene. Behind it are Big Ben and a red telephone box... Above the character, large, puffy knitted text reads 'MERRY CHRISTMAS'. Below the character, smaller knitted text reads 'GET LIT'." |
Prompts in the same conversation:
Prompt 2: “Now, based on that previous image, keep the same cute knitted style and the character wearing the jumper. But translate all the text into Japanese. Change the scene from London to a snowy knitted Tokyo street with a pagoda and glowing paper lanterns instead of Big Ben.”
Prompt 3: “Now, based on the first English image again, translate the text into French. Change the scene to a snowy knitted Paris street with the Eiffel Tower and a small cafe instead of Big Ben. Keep the same character and knitted style.”
Outputs:
Step 3: The Story (Nano Banana Pro + Veo + Gemini TTS + Lyria)
Now let’s generate a storyboard of our character returning home for the holidays and used Veo to turn it into a stop-motion film.
First generate frames for our video which we will then use as start frames for Veo 3.1.
1 reference image + prompt in the same conversation
 | “Generate a wide shot of the cute knitted character walking down a path made of white yarn snow. The character is carrying a vintage suitcase that looks pixelated (8-bit style). It is snowing heavily. The lighting is cinematic twilight blue. The character is wearing the 'Get Lit' jumper. Maintain the 3D isometric wool texture.” |
Output
Prompt 2
“Now show the character standing in front of a cozy cottage made entirely of cable-knit wool textures. Warm yellow light is glowing from the windows. The character is reaching for the door handle. Keep the same 8-bit suitcase and 'Get Lit' jumper. The vibe is welcoming and warm.”
Output
Prompt 3
“Move inside the cottage. Show the character kneeling down to place a wrapped gift under a large knitted Christmas tree. The gift wrap has a retro 8-bit pixel pattern on it. The tree is decorated with glowing LED lights that match the jumper. Focus on the character and the gift.”
Output
Prompt 4
“Final shot. The character is now sitting in a comfy armchair made of yarn, holding a mug of hot cocoa. The fireplace is crackling with 'pixelated' fire (orange 8-bit blocks mixed with wool). The atmosphere is peaceful and happy. The 'Get Lit' jumper lights are glowing brightly.”
Output
Veo 3.1 prompts + 1st frame
 1st frame | A stop-motion animation scene. The knitted character with the pixel suitcase walks down the snowy path. The character moves with a stiff, jerky motion, like an animated puppet. The snow falls in clumpy, wool-like pieces. The camera has a slight, deliberate jerk as it follows the character. The entire scene has a low-frame-rate, tactile, handcrafted feel. Any text is completely rigid and static.” |
 1st frame | “Stop-motion animation. The knitted character opens the cottage door and enters. The movement is jerky and stiff, frame-by-frame. The warm light from the windows flickers slightly, like a physical light source. The text remains completely frozen in place. Low frame rate style” |
 1st frame | Stop-motion animation style. The knitted character kneels and stiffly places the pixelated gift under the tree. The action is jerky and deliberate. The lights on the knitted tree blink in a simple, on-off pattern. The camera angle is fixed and rigid. The text is completely static. The animation is choppy and tactile. |
 1st frame | A stop-motion scene. The knitted character sits in the armchair, holding the mug. The pixelated fire in the fireplace has a jerky, blocky animation, like physical pieces moving. The character's jumper lights have a slow, stiff blink. The camera is completely still. The text is rigidly fixed in the frame. The entire scene feels like a low-frame-rate physical model |
Voiceover prompt for Gemini TTS
Voice style instructions You are a gentle, warm storybook narrator reading a cozy holiday tale. Your voice is soft, smiling, and cute. Speak slowly and melodically, pausing where indicated to let the music swell. | Voice text The weather outside... is looking a bit frightful... ...but the welcome ahead... is simply delightful. He is finally here... to deliver the spark... ...so we can all Get Lit... this holiday season |
Lyria music prompt
“A bright, happy, and bouncy orchestral holiday track. Features playful pizzicato strings and a cute celesta melody that sounds like a toy shop. Rhythmic sleigh bells provide a festive beat. Subtle, sparkling 8-bit chiptune arpeggios play in the background to match a pixel aesthetic. The mood is optimistic, energetic, and adorable. The tempo is lively.”
Then I used video editor to stitch together the individual clips, music and voiceover.
Final output
Step 5: Vibe coding the store (Google AI Studio)
Finally, we needed a place to sell the jumper. I went into Google AI Studio, uploaded the brand guidelines and the jumper, and asked for a functional app.
 |  | "Build a modern, responsive web application for a fashion brand called 'Get Lit'. Use the color palette from this uploaded brand guideline image. The site should have a 'Virtual Try-On' feature where a user can upload their own photo, and the app uses an API to overlay this specific jumper onto them. Write the code in React." |
Watch the full session here 👇
Chapters:
00:00-18:36 Intro + Nano Banana Pro community posts
18:36-26:11 What is Gen media & focus on Nano Banana Pro key missed features
26:11-41:56 Get Lit jumper brand ecosystem
41:56-50:29 Different types of creative personas & building creative products
50:29-59:32 Q&A+ Next steps
Thank you for reading and I hope this was helpful!
Lastly, I just want to say that I am beyond grateful for this community and for all your support you have given my past few months! It’s been so fulfilling to be able to share my learnings with you all and getting inspiration from you on what to test and experiment with next!
I am putting together my plans for 2026, and really want to do more video content and potentially a community page or something of that sort. If there is a way you would like to interact with me, I would love to hear (ping me on Linkedin or reply to this email).
Khulan