Four Scripts, 39 Products: How I Turned One Prompt Engine Into an Assembly Line

Description and use

The author built four scripts that automate CSV generation, upload tracking, product packaging, and Markdown‑to‑PDF conversion, enabling efficient operation on a 1 vCPU/2 GB server and generating sales on Adobe.

Problem

Stock platforms reject CSV uploads when headers do not match their required format, causing wasted time and duplicate uploads.

Outcome

The author produced 1,700 assets, packaged 39 products, and achieved 58 live Adobe listings with a first sale of $0.99 using only a 1 vCPU/2 GB server and minimal third‑party libraries.

Workflow

• Generate platform‑specific CSV with make_metadata.py
• Track upload status with upload_tracker.py to avoid duplicates
• Package assets into zip, cover, and thumbnail with pack_product.py
• Convert Markdown to styled PDF with md2pdf.py
• Run all steps on a single‑process, low‑memory server with memory‑cleanup habits

Evidence

Every stock platform has its own CSV shape. Adobe is 5 columns. Vecteezy is 4. Dreamstime is 15.

Source

I fixed the header by hand, resubmitted, and lost an hour I'll never get back.

Source

The fix is boring: one script that knows the three headers and writes them exactly.

Source

A re‑run only ever touches what's marked pending. It's not clever. That's the point.

Source

Cover rendering is the only place Pillow gets involved.

Source

Everything above runs on a machine with 1 vCPU and 2 GB RAM, no GPU.

Source