Office tea bar overtime sku
Agent Skill system for turning ecommerce product briefs, selling points, and reference assets into validated AI video-generation SKU Skill packages.
npx -y skills add HXZ09845/ecommerce-sku-skill-builder --skill office-tea-bar-overtime-skuAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Use this anonymized ecommerce SKU Skill when turning an office overtime tea-bar appliance brief, selling points, and reference assets into Seedance-style shot contracts, prompt plans, take reviews, and repair rules.
SKILL.md
1.9 KB, as published. Nobody here has run it
Office Tea-Bar Overtime SKU Skill
This is an anonymized output sample generated from a real ecommerce video workflow. It is included to show the package shape that sku-skill-builder can produce. It does not include raw product media, brand names, local paths, or unpublished generated videos.
Scope
Use this Skill for a 9:16 office-overtime product video where a freestanding hot-water appliance helps a desk worker avoid leaving the workflow for water or a hot drink.
Required Reading
Load only what the task needs:
references/product-rules.mdfor product-state and reference-role rules.references/prompt-plan.mdfor Unit schedule and prompt contracts.references/gotchas.mdfor known failures and repair levers.references/trigger-tests.mdfor behavior expectations.
Operating Steps
- Confirm the available source materials and mark missing assets as pending.
- Keep the product position, category, and major silhouette stable across all Units.
- For each Unit, identify one selling point or bridge role.
- Assign each reference asset a role: identity, environment, detail, motion, person, or result-state.
- Write one visible beat and one
Stop whenendpoint per Unit. - Move exact brand text, numeric wattage, link text, screen text, and claims to VO, subtitles, or post-production.
- After generation, fill a take review before changing prompts.
Output Contract
Return:
- source confirmation;
- Unit schedule;
- reference role table;
- Seedance-style prompt plan;
- take-review table;
- repair rules;
- validator result from
scripts/validate_case.pywhen a prompt-plan file is available.