✦ Studio pick
TikTok · 9:16
Rendered by this Skill
✦ Studio pick
Handmade Stage Film v1.0.0
Your campaign photos as a stop-motion stage show: the stills come to life, props cut from them fly around as paper puppets, hand-written captions tell a little story, and it ends on your wordmark and product.
No ratings yet 0 downloads
- Format
- Ad creative · 9:16
- Length
- 0:18
$2.99Buy this Skill
Paste this into your agent. It connects the CLI if needed, then installs the Skill.
Give it the stills from one shoot. It extends each one to 9:16, brings it to life with a few stiff, deliberate movements, keeps faces, products and printed labels exactly as shot, and plays it all back like a handmade stage show: stepped motion, paper puppets cut from your photos, torn-paper captions, scribbled tags with arrows and a music-box waltz, ending on your wordmark, your product and where to buy it.
Updated Nov 3, 2026
What you get
- A 15–20-second brand film from your own campaign stills, with captions, puppets, sound and a last card, as an H.264 MP4 in 9:16
- The 9:16 keyframes, the clips and every cut-out prop as files, plus a poster frame
- An editable JSON file: change the captions, tags, puppets, timing or last card and re-render without paying for new footage
Good fit
- Brands with an art-directed shoot: seasonal drops, limited editions and collabs
- Food and drink, fashion, beauty and lifestyle
- Agencies turning a key-visual campaign into social video
Not for
- Brands without their own photos (it builds on your shoot, it doesn't invent one)
- Talking-head, UGC or testimonial ads
- Fast, realistic action footage
About this Skill
You say
/handmade-stage-film Turn our seven Halloween campaign photos into an 18-second brand film for TikTok: Halloween night, nothing on TV, a ghost climbs out of the screen and goes after her Magic Pretty Water (gardenia, berries, yogurt), and one sip later she's sitting on a giant witch's hat. End on the Molly Tea wordmark, the cup, "Halloween limited" and "Now in US · CA · UK · TH · AU".
Halloween night, nothing on: a girl with a black bob sits inside a painted cardboard sofa, pink drink in hand, and clicks the cardboard remote. The cardboard TV switches on into static, a little ghost dances on the screen, then peels off it and floats away. It hovers by her cup ("It wanted a sip."), lands on a checkered table where a giant fork stabs at a plate of painted berries ("Gardenia. Berries. Yogurt."), and one sip later she's sitting on a giant witch's hat while stars pop on around her ("…full magic."). The cup gets its close-up with a scribbled "Magic Pretty Water", and the last card has the Molly Tea wordmark, the cup, the ghost, "Halloween limited" and where it's on sale. Every frame comes from the brand's own seven campaign photos.
How it works
- Brief: your agent asks for the stills from one shoot (four to eight), the product name, a line, a note and the CTA, then writes a short script: the order of the shots, a little plot, what moves in each one, which props become puppets, and the captions. You see the script and the cost before anything is generated. You also need an API key for Nano Banana 2 and Kling 2.5 Turbo; a film of seven stills costs about $2.30 in model usage.
- The footage: each still is extended to 9:16 with your original pixels laid back on top, and Kling 2.5 Turbo brings it to life with one or two small movements. Props are cut out of the photos to fly around as paper puppets, and any label the video smudged is laid back sharp from your photo.
- The film: the stepped stop-motion playback, the puppets, the torn-paper captions, the scribbled tags, the music-box waltz, the sound and the last card are rendered on your machine in any language, as a 9:16 MP4.
Tips
- Write the clip prompts tight: one or two small actions, a deadpan face, the product held still. Big gestures smudge faces and labels.
- Give the story one traveller, like the ghost here: a puppet cut from one photo that turns up in the next shots ties the stills into one film.
- Cut puppets from your photos, not new art: the same paint, paper and light make them belong.
- Keep captions to three or four words; the pictures tell most of it.
- Give the product one close-up with a tag, and keep the last card for the wordmark, the product and where to buy it.
What you need before you start
- You bring
- Four to eight stills from one shoot, the product name, a short line, a note like "Halloween limited" and your CTA. A wordmark printed on the photos is keyed out for the last card. You also need an API key for Nano Banana 2 and Kling 2.5 Turbo; a film of seven stills costs about $2.30 in model usage.
- Dependencies
- Node.js 20 or newer — required · Google Chrome (or CHROME_PATH) — required · An API key for Nano Banana 2 and Kling 2.5 Turbo — required · FFmpeg — bundled via ffmpeg-static after npm install
- Outside costs
- Model usage, billed to your own API key: $0.08 for each still extended to 9:16 with Nano Banana 2, $0.35 for each 5-second Kling 2.5 Turbo clip, and a fraction of a cent for each cut-out (October 2026 prices). A film of seven stills with five clips is about $2.30. The reel, the puppets, the captions, the sound and the final MP4 are made free on your machine.
- Last tested
- v1.0.0 —
User reviews
No reviews yet. Install the Skill and be the first to rate it.
FAQ
What do I need?
Four to eight stills from one shoot and an API key for the two models the Skill calls: Nano Banana 2 extends each still to 9:16 and Kling 2.5 Turbo brings it to life. You set the key in your own terminal, and the Skill's setup step shows where to get it. Model usage is billed to your key: about $2.30 for a film of seven stills at October 2026 prices.
Will my photos be changed?
No. The model only paints the extra space above and below each still, and your original pixels go back on top. When a clip smudges small printed text, the sharp label from your photo is laid back on every frame.
Where do the puppets come from?
From your photos: a character on a screen, the stars on a prop, a berry on a plate, your product in a hand. Each is cut out and moved around in steps, like a paper puppet.
Is there sound?
Yes. An original music-box waltz and a sound for every moment: paper for the captions, a pen for the tags, pops, whooshes, a remote's click, a TV switching on, a sip and a chime at the end, all generated from scratch, so there's nothing to license.
Can I post it on Xiaohongshu?
Yes. The Skill shows how to crop a 3:4 version from the 9:16 film.
