video-ad-specs

作者: skills-101

針對 TikTok、Instagram、YouTube、Facebook、LinkedIn 的影片廣告創作,提供精確的平台特定規格。涵蓋尺寸、時長限制、AIDA框架及字幕要求。用途:影片廣告、社群媒體廣告、付費媒體創意、影片行銷、廣告製作。觸發詞:video ad, social media ad, tiktok ad, instagram ad, youtube ad, facebook ad, linkedin ad, video creative, ad specs, paid media, video marketing, ad production, reels ad, stories ad, pre roll, bumper ad

npx skills add https://github.com/skills-101/superpowers --skill video-ad-specs

Install the belt CLI skill: npx skills add belt-sh/cli

Video Ad Specs

Create platform-specific video ads via inference.sh CLI.

Quick Start

Requires inference.sh CLI (belt). Install instructions

belt login

# Generate a vertical video ad scene
belt app run bytedance/seedance-2-0 --input '{
  "prompt": "vertical video, person excitedly unboxing a product, clean modern room, bright natural lighting, social media ad style, authentic feeling",
  "ratio": "9:16",
  "generate_audio": true
}'

Platform Specifications

TikTok

SpecValue
Aspect ratio9:16 (vertical)
Resolution1080 x 1920 px
Duration5-60 seconds (15-30s recommended)
File sizeMax 500 MB
FormatMP4, MOV
SoundOn by default (design with sound)
Text safe zone150px from all edges
Hook window1 second — first frame must grab attention

Instagram Reels

SpecValue
Aspect ratio9:16 (vertical)
Resolution1080 x 1920 px
DurationUp to 90 seconds (15-30s for ads)
Cover imageSeparate upload, shows in grid
SoundOn by default
Caption areaBottom 20% reserved for text overlay

Instagram Stories

SpecValue
Aspect ratio9:16
Resolution1080 x 1920 px
DurationUp to 15 seconds per segment
Swipe-up/LinkAvailable for ads
Top/bottom14% top and 20% bottom = unsafe for key content

YouTube

FormatAspectDurationSkip
Bumper16:96 seconds exactlyNon-skippable
Non-skippable16:915 secondsNon-skippable
Skippable (TrueView)16:9Any lengthSkip after 5 seconds
Shorts9:16Up to 60 secondsN/A

Resolution: 1920 x 1080 (16:9) or 1080 x 1920 (Shorts)

Facebook Feed

SpecValue
Aspect ratio1:1 (square) or 4:5 (recommended for mobile)
Resolution1080 x 1080 or 1080 x 1350
DurationUp to 240 min (15-30s recommended)
AutoplaySilent — captions are essential
Sound85% of Facebook video is watched without sound

LinkedIn

SpecValue
Aspect ratio1:1 or 16:9
Resolution1080 x 1080 or 1920 x 1080
Duration3 seconds to 10 minutes (15-30s for ads)
ToneProfessional
AutoplaySilent in feed

AIDA Framework for Video Ads

PhaseTimeGoalTechnique
Attention0-3sStop the scrollPattern interrupt, bold visual, question
Interest3-10sKeep watchingState the problem, show relevance
Desire10-20sWant the solutionShow the product/outcome, social proof
ActionFinal 3-5sClick/buy/sign upClear CTA, urgency, offer

Hook Techniques (First 3 Seconds)

TechniqueExample
Bold statement"This tool replaced my entire marketing team"
Question"Why are you still doing this manually?"
Surprising visualUnexpected transformation, before/after reveal
Pattern interruptStart mid-action, unusual angle, bright color
Social proof"2 million people switched to this"
Pain point"If you hate [common frustration], watch this"

Creating Video Ads

Vertical (TikTok, Reels, Stories, Shorts)

# Hook scene (0-3s)
belt app run google/veo-3-1-fast --input '{
  "prompt": "vertical 9:16 video, close-up of hands struggling with tangled cables and messy desk, frustrated energy, shaky handheld camera, authentic social media style, bright lighting"
}'

# Solution reveal (3-15s)
belt app run bytedance/seedance-2-0 --input '{
  "prompt": "vertical video, smooth product reveal, clean wireless charging station on minimalist desk, satisfying organization transformation, bright modern room, social media ad aesthetic",
  "ratio": "9:16",
  "generate_audio": true
}'

# Add voiceover
belt app run falai/dia-tts --input '{
  "prompt": "[S1] Stop wasting time with this mess. This one product changed my entire setup. Everything charges. Everything is organized. Link in bio."
}'

# Merge video + audio
belt app run infsh/video-audio-merger --input '{
  "video": "solution-reveal.mp4",
  "audio": "voiceover.mp3"
}'

# Add captions (critical for silent autoplay)
belt app run infsh/caption-videos --input '{
  "video": "ad-with-audio.mp4",
  "caption_file": "captions.srt"
}'

Square (Facebook, LinkedIn Feed)

belt app run google/veo-3-1-fast --input '{
  "prompt": "square 1:1 video, professional person at desk discovering a new software tool, laptop screen showing clean dashboard, natural office lighting, corporate commercial style, satisfied expression"
}'

YouTube Bumper (6 Seconds)

# 6-second bumper: one message, one visual, one CTA
belt app run google/veo-3-1-fast --input '{
  "prompt": "6 second product ad, quick montage of a sleek app being used on phone, fast cuts, modern, energetic, brand logo reveal at end, punchy and dynamic, wide 16:9"
}'

# Keep it tight
belt app run falai/dia-tts --input '{
  "prompt": "[S1] Your reports. Automated. Try DataFlow free."
}'

Captions Are Mandatory

85% of Facebook and 40%+ of Instagram video is watched on mute.

Caption Best Practices

RuleReason
Always add captionsSilent viewing is the default on most platforms
Large, readable fontSmall text is invisible on mobile
High contrastWhite text with dark outline/background
Centered or bottom-thirdStandard viewing position
Max 2 lines at a timeMore text = can't be read fast enough
Key words in bold/colorDraws eye to important words
# Generate captions from audio
# (create SRT file from your script, then burn in)
belt app run infsh/caption-videos --input '{
  "video": "ad-video.mp4",
  "caption_file": "ad-captions.srt"
}'

Ad Structure Templates

Testimonial Ad (15-30s)

TimeContent
0-3sCustomer states the problem they had
3-15sHow they discovered and tried the product
15-25sThe specific result they achieved
25-30sProduct name + CTA

Demo Ad (15-30s)

TimeContent
0-3sThe problem (text or visual)
3-20sProduct demo showing the solution
20-25sKey result/benefit
25-30sCTA + offer

Before/After Ad (15s)

TimeContent
0-3s"Before" state (messy, slow, frustrating)
3-5sTransition / product introduction
5-12s"After" state (clean, fast, satisfying)
12-15sCTA

Common Mistakes

MistakeProblemFix
No hook in first 1-3sViewer scrolls pastOpen with pattern interrupt
Landscape video on TikTok/ReelsLetterboxed, looks amateurUse 9:16 for vertical platforms
No captionsMost viewers watch silentAlways add captions
CTA too lateViewers already leftClear CTA within last 5 seconds
Too long for platformForced skip or dropoutMatch platform duration norms
Same ad for all platformsWrong specs, wrong toneCreate platform-specific versions
Logo in first 3sFeels like a commercial, gets skippedSave branding for the end
Text in unsafe zonesCut off by platform UICheck safe zone per platform

Checklist

  • Correct aspect ratio for target platform
  • Hook in first 1-3 seconds
  • Captions added (readable, high contrast)
  • CTA clear and within final 5 seconds
  • Duration matches platform norms
  • Text outside platform unsafe zones
  • Audio designed for both sound-on and sound-off
  • Platform-specific version (not one-size-fits-all)

Related Skills

npx skills add inference-sh/skills@ai-video-generation
npx skills add inference-sh/skills@video-prompting-guide
npx skills add inference-sh/skills@text-to-speech
npx skills add inference-sh/skills@prompt-engineering

Browse all apps: belt app list

來自 skills-101 的更多技能

agent-browser
skills-101
透過 inference.sh 為 AI 代理提供瀏覽器自動化。瀏覽網頁、使用 @e 引用與元素互動、截取螢幕截圖、錄製影片。功能:網頁抓取、表單填寫、點擊、輸入、拖放、檔案上傳、JavaScript 執行。用途:網頁自動化、資料提取、測試、代理瀏覽、研究。觸發詞:瀏覽器、網頁自動化、抓取、導覽、點擊、填寫表單、截圖、瀏覽網頁、playwright、無頭瀏覽器、網頁代理、上網、錄製影片
agent-tools
skills-101
透過 inference.sh CLI 執行 AI 應用程式 - 影像生成、影片創作、LLM、搜尋、3D、Twitter 自動化。模型:FLUX、Veo、Gemini、Grok、Claude、Seedance、OmniHuman、Tavily、Exa、OpenRouter 等眾多模型。適用於執行 AI 應用程式、生成影像/影片、呼叫 LLM、網路搜尋或自動化 Twitter。觸發詞:inference.sh、infsh、ai model、run ai、serverless ai、ai api、flux、veo、claude api、image generation、video generation、openrouter、tavily、exa search、twitter api、grok
ai-avatar-video
skills-101
透過 inference.sh CLI 建立 AI 虛擬人偶與說話頭影片。建議使用:P-Video-Avatar(最快、最便宜、內建 TTS)。其他選項:OmniHuman、Fabric、PixVerse。音訊:Inworld TTS-2(100 多種語言、角色情緒引導)、ElevenLabs、Kokoro。功能:音訊驅動虛擬人偶、文字轉虛擬人偶、唇形同步影片、說話頭生成、虛擬主持人、UGC 內容。適用於:AI 主持人、解說影片、虛擬網紅、配音、行銷影片、UGC 廣告、遊戲虛擬人偶、...
ai-image-generation
skills-101
透過 inference.sh CLI 使用 GPT-Image-2、FLUX、Gemini、Grok、Seedream、Reve 及 50 多種模型生成 AI 圖像。模型:GPT-Image-2、FLUX Dev LoRA、FLUX.2 Klein LoRA、Gemini 3 Pro Image、Grok Imagine、Seedream 4.5、Reve、ImagineArt。功能:文字轉圖像、圖像轉圖像、修復、LoRA、圖像編輯、放大、文字渲染。用途:AI 藝術、產品模型、概念藝術、社群媒體圖形、行銷視覺、插圖。觸發詞:flux、image generation、ai image、text to...
ai-video-generation
skills-101
使用 Google Veo、Seedance 2.0、HappyHorse、Wan、Grok 及 40 多種模型,透過 inference.sh CLI 生成 AI 影片。模型:Veo 3.1、Veo 3、Seedance 2.0、HappyHorse 1.0、Wan 2.5、Grok Imagine Video、OmniHuman、Fabric、HunyuanVideo。功能:文字轉影片、圖片轉影片、參考素材轉影片、影片編輯、嘴型同步、虛擬人物動畫、影片畫質提升、擬音音效。適用於:社群媒體影片、行銷內容、解說影片、產品示範、AI 虛擬人物。觸發條件:影片生成、AI 影片、...
app-store-screenshots
skills-101
App Store 和 Google Play 截圖製作,符合精確的平台規格。涵蓋 iOS/Android 尺寸、圖庫排序、裝置模型和預覽影片。適用於:App Store 優化、ASO、應用程式截圖、應用程式預覽、Play Store 商店頁面。觸發詞:App Store 截圖、ASO、App Store 優化、Play Store 截圖、應用程式預覽、應用程式商店頁面、iOS 截圖、Android 截圖、App Store 圖片、應用程式模型、裝置模型、應用程式圖庫、商店頁面
character-design-sheet
skills-101
AI生成圖像中的角色一致性,透過參考表與LoRA技術達成。涵蓋多角度視圖、表情表、配色方案與風格一致性技巧。適用於:角色設計、遊戲美術、插畫、動畫、漫畫、視覺小說。觸發詞:角色設計、角色表、角色一致性、角色參考、多角度視圖表、表情表、角色美術、一致角色、角色概念、參考表、角色創作、原創角色設計、...
competitor-teardown
skills-101
結構化競爭分析,包含功能矩陣、SWOT、定位圖及使用者體驗審查。涵蓋研究框架、價格比較、評論挖掘與視覺化成果。適用於:市場研究、競爭情報、投資者簡報、產品策略、銷售賦能。觸發詞:競爭者分析、競爭分析、對手拆解、市場研究、競爭情報、SWOT分析、競爭者比較、市場格局、競爭者審查、競爭...