- Home
- Skills
- Image & Video
- Nano Batch Studio
Nano Batch Studio
Styling product photos into lifestyle scenes one at a time is slow, and the results drift from shot to shot.
$10
Works with the AI Agents you already use
Nano Batch Studio
Example session with this skill installed
I've put 4 furniture product photos in the photos folder (two lounge chairs, a recliner and a round coffee table). Place each one in a minimalist Scandinavian living room: warm natural light, light oak floors, soft neutral textiles, editorial magazine look. Keep each product exactly as it is and fully visible. 2K, 3:2, JPG.
- Read your context and instructions
- Compiled the nano batch studio
Summary
Four product photos styled into Scandinavian living-room scenes with Nano Banana Pro at 2K. All four succeeded, and each output kept its original filename.
Plan (shown before generating)
| Setting | Value |
|---|---|
| Mode | scenes (one scene per photo) |
| Model | nano-banana-pro |
| Size | 2K, 3:2 |
| Images | 4 |
| Output folder | outputs/ |
You confirmed, so I ran the batch.
Result
| Input | Output | Size |
|---|---|---|
lounge-chair-1.webp | outputs/lounge-chair-1.jpg | 2528 × 1696 |
lounge-chair-2.jpg | outputs/lounge-chair-2.jpg | 2528 × 1696 |
recliner.webp | outputs/recliner.jpg | 2528 × 1696 |
coffee-table.webp | outputs/coffee-table.jpg | 2528 × 1696 |
4 created, 0 failed.
Each product sits on a light oak floor with a textured rug, sheer-curtained windows and warm side light. Shapes, materials and colours match the source photos.
Next steps
- Want a different room style for the same products? Give me a new scene description and I'll re-run; existing files won't be overwritten.
- For print, re-run any favourites at 4K.
- To furnish the rooms with only your own products instead of model-invented decor, point me at a folder of your catalogue photos.
Connects securely to your tools. The creator never sees your data.
About this skill
Styling product photos into lifestyle scenes one at a time is slow, and the results drift from shot to shot. Nano Batch Studio applies one scene description to a whole folder of product photos and saves each result under the original filename, ready to drop back into a catalogue.
What it does
- Photo scenes Places each product photo into its own styled scene. One photo in, one image out, original filename kept.
- Product list Generates images from a CSV of product names, with no photos needed.
- Combine Blends several product photos into one room and makes as many variations as you ask for.
- Furnish with your own catalogue Point it at a folder of your products and the model may use only those to dress the scene, instead of inventing furniture and decor.
- Build a library from a web page Downloads product photos from a page you have rights to, skipping logos, icons and thumbnails.
- Safe batching Always plans first and asks before generating more than a few images. Files are never overwritten. An account problem such as a rejected key stops the batch at the first failure instead of failing every image.
How it works
- Tell your agent where the photos are and what the scene should look like.
- It runs
planand shows you how many images it will make, with which model and size. - After you confirm, it generates the images and reports where they are and anything that failed, with a plain-English reason.
Models
Nano Banana Pro, Nano Banana 2, Nano Banana 2 Lite, GPT Image 2, GPT Image 2.5 Flare and GPT Image 2.5 Sunburst. Sizes 1K, 2K and 4K, in any common aspect ratio. Runs on Agensi's image tools, so no API key is needed; if you have your own Gemini or OpenAI key, it can use that instead.
Use cases
- Lifestyle images for a furniture or homeware catalogue
- One product shown across several room styles
- Product shots where the background uses only your own range
- Fast 1K drafts, then 2K or 4K finals of the ones you like
Known limitations
- Generation runs through image models, so a product can occasionally drift in shape or colour. Fewer library items, an explicit instruction to keep the product unchanged, or GPT Image 2.5 Sunburst all help.
- Web-page library fetching reads plain HTML; shops that load images with JavaScript return nothing, so save those images to a folder instead.
- Needs Python 3.10+ on the machine your agent runs on.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Video Skills.
Needs Python 3.10+. Up to 13 library items per scene. Asks before generating large batches, since each image costs money. Products can occasionally drift in shape or colour; fewer library items or GPT Image 2.5 Sunburst help. Web-page library fetching only reads plain HTML pages.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 6 days ago
- Software
Python 3.10+
- Optional
- Your own Geminior
OpenAI API key (only to bypass Agensi tools)
Permissions
Shell: runs the bundled Python CLI. Read files: your product photos, product CSV and optional library folder. Write files: saves generated images to the output folder (never overwrites). Network: image model calls, and only when you ask it, fetching product photos from a web page. Environment variables: reads GEMINI_API_KEY / OPENAI_API_KEY only if you choose to use your own key; keys are never accepted as arguments.