Communitygithub.com

Akira-TL/akira-video-skills

Establishes product facts and advertising boundaries for brand short films, product-embedded narratives, commercials, or competition videos; use when a video needs to accurately represent the official features, specifications, structure, usage, brand assets, or product-story relationship of a specific product.

What is akira-video-skills?

This skill provides a structured framework for AI agents to handle brand and product references in video production with accuracy and integrity. It is designed for scenarios such as creating short brand films, embedding products in narratives, or developing competition entry videos, ensuring all product-related content is factual and aligned with official materials. The skill supports platforms like Cursor, Codex, and Claude Code, and operates by reading from dedicated reference files (`PRODUCT-TRUTH.md` and `PRODUCT-SHOTS.md`), applying rules that distinguish confirmed facts from speculation, respecting brand safety, and limiting end cards to essential information. Its primary benefit is preventing the AI from inventing unverified product capabilities or misrepresenting the product's role in the story, thus maintaining credibility and legal safety.

Works with~Claude Code~Codex CLI~Cursor
npx skills add https://github.com/Akira-TL/akira-video-skills/tree/HEAD/skills/video/video-advertising

Ask in your favorite AI

Open a new chat with this agent skill pre-loaded.

Documentation

Video Advertising

This Skill is an optional domain capability, used only when the current video genuinely involves a brand or specific product. It does not turn every video into an advertising workflow.

When establishing or updating specific product facts, read references/PRODUCT-TRUTH.md. When directly displaying, generating, or using real product footage, then read references/PRODUCT-SHOTS.md.

1. Product Facts

Start by creating a fact file for the current product based on official materials provided or explicitly authorized by the user. It is recommended to co-locate it with product assets, e.g., video/materials/products/PROD01_truth.md.

At minimum, distinguish between:

  • Features and specifications that are officially confirmed.
  • Structure and usage instructions.
  • Capabilities that can safely enter the plot/scene.
  • Content that requires cautious expression or still needs manual confirmation.
  • Content that current materials cannot prove.

Keep unknowns as unknowns. Do not automatically apply product line capabilities to a specific model, do not interpret marketing language as technical mechanisms, and do not add unverified capabilities (automatic recognition, automatic control, speed, distance, temperature, etc.) for the sake of the plot.

2. Official Assets vs. Generation References

Clearly distinguish between official product images, logos, model numbers, structural materials, and real usage assets from AI-generated references. Product industrial structure takes priority over stylization; when there is no reliable back or structural angle, do not treat AI guesses as official structural evidence.

3. Product and Plot

In narrative advertisements, the product should solve a problem that has already been established, and serve a decision the character has already made—rather than appearing abruptly at the end or completing the character's personality transformation.

Prioritize highlighting one core selling point; supplementary selling points should only be added when genuinely necessary. If the story's resolution is completely unaffected when the product is removed, re-examine the product's relationship to the plot.

4. Brand Safety and End Cards

When the product is not a fantasy prop, maintain its real industrial identity. Do not arbitrarily turn real brand products into villains causing harm, magical items, or personified objects with a completely different structure.

By default, the end card should only include what is truly necessary for the current deliverable: brand, model number, one main message, and at most 1–2 confirmed features/selling points. The story handles emotion; the end card ensures the audience knows what the product is and what problem it solved.

Do not default to cramming six or seven selling points, complex specifications, story theme summaries, or large amounts of promotional slogans onto one main card. Prices should only appear when the current brief / official deliverable explicitly requires them and the fact source is reliable.

Precise logos, text, model numbers, and specifications should be completed deterministically using official files during post-production for the entire video.

Completion Criteria

The product's features, structure, and promotional expressions in the current video can all be traced back to clear official sources; the advertising narrative and shots do not cross these factual boundaries for creativity.

Individual skills in this repo

This repo contains 12 individual skills — each has its own dedicated page.

Akira-TL/akira-video-skills

为 Runway 当前视频模型编译和检查生成提示词;当 video-generation 需要使用 Runway(当前主线 Gen-4.5)的 Text to Video 或 Image to Video,并需要 Runway 专用动作、摄影和时序表达时使用。

Akira-TL/akira-video-skills

为 ByteDance / BytePlus Seedance 当前视频模型编译和检查提示词;当 video-generation 需要使用 Seedance 2.5 的参考素材、时间线、同步声音、视频编辑、延长或多镜头能力时使用。

Akira-TL/akira-video-skills

为 Google Veo 视频生成编译和检查提示词;当 video-generation 需要使用 Veo 的 Text to Video、Image to Video、首尾帧、参考元素或同步声音能力时使用,并先核验具体 Google 使用入口与 model 的当前支持范围。

Akira-TL/akira-video-skills

进入并持续推进一个 AI 视频制作项目;读取或建立 VIDEO.md,根据当前视频脚本、复用素材、镜头、生成结果和整片后期状态,按需路由到 canonical 视频专业 Skill,而不是把所有制作规则塞进一个总提示词。

Akira-TL/akira-video-skills

设计并维护 AI 视频中跨镜头复用的角色声音、音色参考、对白声音要求与整片声音关系;当同一角色需要跨镜头保持声音一致,或需要规划语音参考、环境声、音效、音乐与画面的职责边界时使用。

Akira-TL/akira-video-skills

设计 AI 视频中的角色与场景视觉方案;当脚本只说明“需要一个什么样的人/场景”,但还没有形成可稳定生成的外形、服装、材质、色彩、空间、光线和视觉识别系统时使用。

Akira-TL/akira-video-skills

组织整片级剪辑、合成、全片声音和最终成片;当多个已采用镜头需要进入 Premiere Pro、After Effects、DaVinci Resolve 等后期工程,或需要整片字幕、Logo、音乐、混音、比例版本和最终导出时使用。

Akira-TL/akira-video-skills

为 AI 图片或视频生成准备可直接使用的提示词与一次性生成包,并在用户把生成结果带回项目后归档;当需要从正式视频脚本、复用素材和 SHOT.md 编译生成内容、选择参考素材、约定返回文件名或清理临时包时使用。

Akira-TL/akira-video-skills

规划、创建、整理和维护多个镜头会重复使用的视频复用素材;当需要人物/场景/道具/产品提示词、参考图、官方素材、身份与状态参考或其他可重复引用内容时使用。

Akira-TL/akira-video-skills

审核 AI 生成的人物/场景等复用素材或镜头生成结果,判断采用、修改提示词、重生成还是留给后期修复;当生成结果返回项目、需要检查身份/结构/连续性/动作/摄影/产品真实性或选择生成结果时使用。

Akira-TL/akira-video-skills

把小说、章节、故事梗概、品牌 brief 或其他上游内容转换成当前视频自己的脚本层;当需要编写或修订 SCRIPT.md、视频人物表演定义、场景定义、对白或镜头总表时使用。

Akira-TL/akira-video-skills

把 SHOTS.md 中的镜头总览展开成可直接制作的单镜头定义;当需要创建或修改 video/shots/<shot-id>/SHOT.md、镜头连续性、摄影、动作、表演、声音或镜头专属生产约束时使用。

Related Skills