TikTok Shop Product Video Maker

Specific Problem: From Product Facts to Executable Video Plans

When merchants or creators need to produce TikTok Shop product videos, they often face a core gap: having product images, basic selling points, and target market info, but struggling to transform these scattered facts into a shootable video plan. This includes opening hooks, voiceover scripts, shot beats, subtitle prompts, and localized copy—all while ensuring every claim strictly adheres to merchant-provided facts, avoiding speculation. Traditional methods may rely on manual creativity or generic templates, often overlooking market language adaptation, factual consistency, or shoot feasibility, leading to plans that are either impractical or disconnected from the product's real advantages.

This skill addresses this problem: based on confirmed product facts, it systematically generates a structured video plan, enabling sellers or creators to start filming quickly while preventing factual errors or overpromising.

How It Works: Core Capabilities and Key Steps

The skill's core is fact-driven video plan generation, not video production itself. It operates through the following steps:

• Input Collection: The hard input is confirmed merchant facts, including product name, category, selling points, price, audience, target market, etc. The skill reuses existing conversation info, only probing key items that might change the plan (e.g., target market, desired audience action). Inputs must be processed in the target market language—for instance, outputting English for US audiences.
• Opening Hooks and Script Generation: Starting from product facts, it generates 3 distinct opening hooks based on real advantages, selecting one as the main hook. Then, it writes a 20–45 second voiceover script sentence by sentence, ensuring each claim corresponds to a provable visual. The script follows a structure of hook → proof or demo → quote → call to action.
• Shot and Subtitle Prompts: Each voiceover segment maps to a shot beat (describing camera movement or visuals), paired with a subtitle prompt. All content is confined to merchant facts—for example, product images can only describe visible elements, not serve as evidence for unmentioned facts.
• Localization and Delivery: It generates localized titles, hashtags, and calls to action. The delivery includes the main script, two backup openings, and a fact checklist for merchant confirmation. All wording is screened: no result promises, no absolute statements, no health/medical/efficacy claims (if hit, rewrite instead of adding disclaimers).

A key capability is strict adherence to the fact chain: from briefing → shot-by-shot plan → voiceover → subtitles, each step anchors to merchant-provided data, avoiding fabricated details. The skill does not generate videos, audio, or list products—it only outputs plan files.

Applicability Boundaries and Considerations

This skill is explicitly a plan-working tool, with the following limitations:

• No Video Production: It provides scripts and shooting guidelines but does not film, record voiceovers, list products, or run ads. For single-image-to-finished-video, use product-video-studio; for AI creator UGC, use douyin-ugc-ad-creator; for voiceover audio, use short-form-voiceover-audio.
• Fact Dependency: All content relies on merchant facts. If key proof points are missing, the skill stops and asks rather than filling with plausible details. Dynamic data like price, inventory, and reviews must come from merchants, not inferred from product images.
• No Result Promises: It does not guarantee virality, sales, or platform approval. The plan only describes how to shoot, not operations or advertising.
• Localization Rules: Language auto-adapts to the target market (e.g., English for the US), with brand terms kept as-is. For revisions, changes in market, facts, or audience require a new briefing, modifying only affected fields.

Usage notes: The skill operates via a bundled MCP client scripts/mcp_client.py, not configuring the host Connector; auto-updates run silently without interrupting current execution. Overall, it's a video plan generator focused on factual consistency, suitable for teams needing fast, accurate planning for TikTok Shop content.