From Prompt To Packaged Product In Under An Hour: The Four Scripts That Made 39 Stock Assets

Description and use

Four scripts automated metadata creation, upload tracking, document conversion, and packaging, enabling the author to prepare 39 ready‑to‑upload assets in an hour on a 1 vCPU, 2 GB RAM server.

Problem

Uploading 39 finished assets to stock channels failed due to CSV header mismatches and risk of double‑posting, preventing the assets from being listed.

Outcome

The four scripts automated metadata creation, upload tracking, document conversion, and packaging, enabling the author to produce 39 ready‑to‑upload stock assets in under an hour on a 1 vCPU, 2 GB RAM server.

Workflow

• Run python make_metadata.py to generate a CSV with the exact header for the chosen platform.
• Use python upload_tracker.py to list pending items and mark items as done after upload.
• Convert markdown documents to PDFs with python md2pdf.py.
• Package assets and generate covers with python pack_product.py.
• Upload the resulting ZIP to the stock channel.
• Maintain a folder layout with assets/, output/, packs/, research/, and scripts/.
• Use a single‑core server with memory‑management practices such as gc.collect() and single‑process rendering.

Evidence

Adobe wants 5 columns. Vecteezy wants 4. Dreamstime wants 15.

Source

The double‑posting fix was the simplest script of the four.

Source

Half my products are documents: guides, pattern sheets, license notes.

Source

The last script bundles everything. pack_product.py zips the files and renders a cover at 1280×720 plus a thumbnail at 600×600.

Source

The whole thing ran on 1 vCPU and 2 GB RAM, so a few habits carried the load: One process at a time. Never fan out image jobs in parallel on a 1-core box.

Source