
video-caption-creation
Create optimized captions and hooks for short-form video (TikTok, Reels, Shorts) using proven hook categories and platform-specific Triple Word Score optimization system.
- Installs
- 61
- Last 24h
- —
- Trending rank
- —
- Security
- Not audited
Install
Works with every agent the skills CLI supports.
npx skills add https://github.com/cdeistopened/skill-stack --skill video-caption-creation- 01
Install
Run the command in your project. The CLI asks which agents to install for.
- 02
Describe the task
Ask your agent for something this skill covers. It loads video-caption-creation automatically when relevant.
- 03
Review the result
The agent follows the SKILL.md steps, scripts and references bundled with the skill.
SKILL.md
The instructions your agent reads when the skill is triggered.
Video Caption Writer
Purpose
This skill creates algorithm-optimized captions and on-screen text for short-form video content using the "Triple Word Score" system to maximize reach and engagement.
Core Philosophy: Content serves the topic first. When you serve the topic correctly, algorithms will serve your content to people interested in that topic.
When to Use This Skill
Use this skill when you need to:
- Create on-screen text hooks for video clips (3-5 options per clip)
- Write captions for short-form video (TikTok, Instagram, YouTube Shorts, etc.)
- Optimize video for algorithm discovery (topic identification in first 10 seconds)
- Generate strategic hashtag sets (10-12, broad to specific)
Do NOT use for:
- Text-only social posts (use
social-content-creationskill instead) - Long-form YouTube videos (separate workflow)
- Podcast show notes (separate podcast workflow)
The Triple Word Score System
Algorithms and humans need to immediately understand your content through FOUR signals:
1. Audio Transcript (MOST IMPORTANT)
- What you say out loud - algorithms auto-transcribe this
- Explicitly say topic words in first 10 seconds
- Repeat core terminology throughout
- Use natural speech patterns with topic-relevant vocabulary
2. On-Screen Text Reinforcement
- Visual overlay that reinforces spoken words
- Lead with topic words in first text overlay
- Use bold, clear terminology
- Don't compete with audio - reinforce it
3. Caption Copy Optimization
- Post description with topic-relevant keywords
- Open with topic-relevant phrases in first sentence
- Include natural keyword variations
- Provide context for algorithm understanding
4. Strategic Hashtag System
- 10-12 hashtags (optimal range)
- Mix broad and specific (#Productivity + #TimeManagement)
- Include audience-specific (#Entrepreneurs)
- Span from general to niche
When all four signals align, algorithms AND humans immediately recognize: "This is for me."
The 3-Step Workflow
Step 1: Generate On-Screen Text Hooks
Goal: Create 3-5 hook options that stop the scroll in first 3 seconds
The McDonald's Test
"Someone who works at McDonald's should be able to read and understand your headline instantly."
- Avoid fancy vocabulary ("optimization paradigm" vs "work smarter not harder")
- Choose accessible language over impressive terminology
- Wider net = better performance
Hook Categories
1. Polarizing Statements
- "Your [perceived weakness] is actually your superpower"
- "Stop chasing [common goal everyone pursues]"
- "[Respected institution] operates like [controversial comparison]"
2. Counter-Intuitive Reveals
- "My first attempt failed. Perfect."
- "The worst experiences are the best teachers"
- "Everyone has genius hiding somewhere"
3. Direct Challenges
- "Never give up on the outliers"
- "Don't judge people by their [surface trait]"
- "Make your audience love [topic] again"
4. Curiosity Gaps
- "Her boyfriend freaked out..." (for dramatic content)
- "The third tip will blow your mind"
- Use carefully - can create confusion if unclear
Generate 3-5 Options
For each video clip, create file: [Clip Name]_Captions.md
Format:
# [Clip Name] - Caption Options
## Source Material
[Brief description of video content, key quote, or main message]
## On-Screen Text Hook Options (First 3 Seconds):
**Option 1:** Your biggest weakness is actually your superpower
- **Framework:** Polarizing Statement
**Option 2:** Stop chasing perfection. Try this instead.
- **Framework:** Direct Challenge
**Option 3:** The one thing experts get wrong about creativity
- **Framework:** Curiosity Gap
**Option 4:** Most advice teaches compliance. This teaches results.
- **Framework:** Contrast/Counter-Intuitive
**Option 5:** 3 reasons your "bad habit" might be genius
- **Framework:** Numbered List + Counter-Intuitive
**RECOMMENDED:** Option 1 (Polarizing) for Short-Form, Option 4 (Contrast) for LinkedIn
Quality Check for Hooks
- Would this make ME stop scrolling?
- Creates curiosity or emotion in 2 seconds?
- Can my grandmother understand instantly?
- Hints at payoff without giving it away?
- Passes McDonald's Test (accessible language)?
Step 2: Create Platform-Specific Captions
Goal: Write optimized captions for each target platform
Short-Form Video Platforms (Identical Treatment)
Applies to: YouTube Shorts, Instagram Reels, TikTok, Facebook Reels
- Character Limit: 2,200 (use 50-150 characters optimal)
- Hashtags: 3-5 (include platform tag: #Shorts for YouTube)
- Tone: Casual, energetic, hook in first few words
- Example: "Your biggest weakness might be your superpower 💡 #Productivity #Mindset #Growth"
Key: Hook immediately, use emojis, keep text scannable
LinkedIn (Separate Strategy)
Platform-Specific Guidelines:
- Character Limit: 3,000 (use 150-300 characters optimal)
- Hashtags: 3-5
- Tone: Professional yet conversational, value-driven
- Format: Hook → insight → CTA (save, comment, share)
- Example: "Your so-called 'weakness' might be teaching you more than any course. Constraints build creativity and resourcefulness. #Productivity #Mindset #GrowthMindset"
Note: For LinkedIn posts, use the social-content-creation skill instead of this skill.
Platform-Specific Caption Template
## Short-Form Video Captions
### YouTube Shorts / Instagram Reels / TikTok / Facebook Reels
Your biggest weakness is actually your superpower 💡
Traditional advice teaches compliance. This teaches results.
#Productivity #Mindset #Growth #Success
---
### LinkedIn (Use social-content-creation skill)
Your so-called "weakness" might be teaching you more than any course.
Constraints force creativity, problem-solving, and resourcefulness. Traditional advice teaches compliance. Real growth comes from embracing what makes you different.
#Productivity #Mindset #GrowthMindset
Reference for detailed platform guidelines: references/platform-guidelines.md
Step 3: Apply Algorithm Optimization
Goal: Ensure content is discoverable and serves the right audience
Topic Identification Checklist
First 10 Seconds (Critical):
- Topic words spoken explicitly in audio
- On-screen text reinforces topic words
- Visual elements support topic (environment, materials, context)
- Energy/delivery matches topic expectations
Throughout Video:
- Core terminology repeated naturally
- Topic focus maintained (not scattered across multiple topics)
- Value delivered on topic promise
Caption & Metadata:
- Caption opens with topic-relevant phrase
- Natural keyword variations included
- Hashtags span broad to specific (10-12 total)
- Description provides context for algorithm
Strategic Hashtag System
Formula: Broad → Mid → Specific → Niche
Example for productivity content:
Broad (1-2): #Productivity #Success
Mid (2-3): #TimeManagement #PersonalDevelopment #WorkSmarter
Specific (3-4): #DeepWork #FocusHacks #MorningRoutine #ProductivityTips
Niche (2-3): #SolopreneurLife #RemoteWorkTips #DigitalNomad
Audience (1-2): #Entrepreneurs #Freelancers
Platform (1): #Shorts (for YouTube), #Reels (for Instagram)
Total: 10-12 hashtags
Common Mistakes to Avoid
❌ Misleading Signal Trap
- Topic says "education" but content is really entertainment
- ✅ Fix: Topic signals match actual content value
❌ Buried Lede
- Topic words don't appear until middle of content
- ✅ Fix: Topic identification happens in first 10 seconds
❌ Scatter Shot
- Trying to cover multiple topics in one piece
- ✅ Fix: Laser focus on one clear topic per piece
❌ Brand First Approach
- Making content about your brand/company
- ✅ Fix: Making content about the topic, featuring your expertise
Output Format
Standard Output File: [Clip Name]_Captions.md
# [Clip Name] - Captions & Optimization
## Video Details
**Duration:** [XX seconds]
**Core Message:** [1-2 sentence summary]
**Topic:** [Primary topic for algorithm]
**Target Audience:** [Who this serves]
---
## On-Screen Text Options (Choose 1)
**Option 1:** [Text]
- Category: [Hook category]
- Platform fit: [Platforms]
**Option 2:** [Text]
- Category: [Hook category]
- Platform fit: [Platforms]
**Option 3:** [Text]
- Category: [Hook category]
- Platform fit: [Platforms]
**RECOMMENDED:** [Which option and why]
---
## Platform-Specific Captions
### TikTok
[Caption text with emojis]
[Hashtags: 3-5]
---
### Instagram Reels
[Caption text with emojis]
[Hashtags: 5-10]
---
### YouTube Shorts
[Caption text]
[Hashtags: 3-5 including #Shorts]
---
### LinkedIn
[Caption text - more professional tone]
[Hashtags: 3-5]
---
### Facebook
[Caption text with emojis]
[Hashtags: 2-5]
---
## Algorithm Optimization
**Topic Keywords (spoken in first 10 seconds):** [List keywords]
**Secondary Keywords (throughout):** [List keywords]
**Hashtag Set (10-12 total):**
- Broad: [2 tags]
- Mid: [3 tags]
- Specific: [4 tags]
- Niche: [3 tags]
**First 10 Second Audio Script:**
"[Exact words that will be spoken - must include topic keywords]"
---
## Quality Checks
- [ ] McDonald's Test passed (accessible language)
- [ ] Topic clear in first 3 seconds
- [ ] On-screen text reinforces (not competes with) audio
- [ ] Platform character limits followed
- [ ] Hashtag counts appropriate per platform
- [ ] Triple Word Score complete (audio, on-screen, caption, hashtags)
Bundled Resources
This skill references:
Detailed Guidelines
references/platform-guidelines.md- Platform-specific rules and best practicesreferences/hook-categories.md- Detailed hook category examples and when to use each
Style & Voice
voice-pirate-wiresskill - Authentic voice principlesanti-ai-writingskill - Humanize captions that sound too polished
Success Metrics
A successful video caption:
✅ Stops the scroll - Hook creates curiosity/emotion in first 3 seconds ✅ Passes McDonald's Test - Accessible to broad audience ✅ Triple Word Score complete - Audio, on-screen, caption, hashtags all aligned ✅ Topic-first - Algorithm knows what this is about immediately ✅ Platform-optimized - Character limits, hashtag counts, voice appropriate ✅ Discoverable - Strategic hashtags span broad to niche
Related Skills
social-content-creation- For text-only social posts (LinkedIn, X/Twitter, Facebook)youtube-title-creator- Pair captions with optimized video titleshook-and-headline-writing- Foundational hook strategies
Common Pitfalls to Avoid
Content Issues: ❌ Jargon or fancy vocabulary in hooks ❌ Giving away the payoff in the hook ❌ On-screen text competes with (instead of reinforces) audio ❌ Generic hooks everyone uses
Algorithm Issues: ❌ Topic words not spoken in first 10 seconds ❌ Too many topics in one video ❌ Hashtags all broad or all niche (not spanning range) ❌ Misleading hooks that don't match content
Platform Issues: ❌ Same caption copy across all platforms ❌ Too many hashtags (LinkedIn, X) or too few (Instagram) ❌ Wrong tone for platform (too casual for LinkedIn, too formal for TikTok) ❌ External links where they hurt reach (Facebook main post)
Version History
- v1.0 (2025-10-27): Initial skill creation
- Triple Word Score system
- 4 hook categories
- Platform-specific guidelines for 6 platforms
- Algorithm optimization checklist
For platform updates or new hook categories, add to references folder and note in version history
FAQ
What does the video-caption-creation skill do?
Create optimized captions and hooks for short-form video (TikTok, Reels, Shorts) using proven hook categories and platform-specific Triple Word Score optimization system.
How do I install video-caption-creation?
Run npx skills add https://github.com/cdeistopened/skill-stack --skill video-caption-creation in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/video-caption-creation/ (personal) or .claude/skills/video-caption-creation/ (project).
Which AI agents can use video-caption-creation?
Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.
Is video-caption-creation safe to install?
No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.
Who maintains video-caption-creation?
It is published by cdeistopened in cdeistopened/skill-stack. It has 61 recorded installs.
More from cdeistopened/skill-stack
youtube-scriptwriting
cdeistopened/skill-stack
Transform raw ideas and brain dumps into polished YouTube scripts through a structured checkpoint workflow. Use when the user wants to write a YouTube script, improve video retention, craft hooks, structure educational or entertainment content, or turn source material (transcripts, notes, research) into a compelling video script. Guides through research, hook writing, story structure, body content, and editing phases.
image-prompt-generator
cdeistopened/skill-stack
Open the skill to read its SKILL.md instructions.
anti-ai-writing
cdeistopened/skill-stack
Open the skill to read its SKILL.md instructions.
amazon-category-research
cdeistopened/skill-stack
Open the skill to read its SKILL.md instructions.
hook-and-headline-writing
cdeistopened/skill-stack
Open the skill to read its SKILL.md instructions.
ghostwriter
cdeistopened/skill-stack
Master the art of converting source material into authentic, human-written content. Transform transcripts, notes, and research into polished prose that avoids AI tells while maintaining distinctive voice. Combines anti-AI writing fundamentals with voice style adaptation.
Popular media & creative skills
hyperframes
heygen-com/hyperframes
Read first to make, edit, animate or render any video with HyperFrames (HTML rendered to video), including launch films, explainers, music videos, charts, maps, vertical clips, title cards, captions and overlays. Holds the render contract, the work loop and what each kind of video needs.
media-use
heygen-com/hyperframes
Get or make any media for a HyperFrames video - music, sound effects, images, icons, logos, voiceover, avatar and generated video, transcripts, cutouts, colour grades and looks - with HeyGen or free local tools, and mix placed audio. Also for footage that looks dark, flat or dull, or should be stylized or hidden.
hyperframes-registry
heygen-com/hyperframes
About 450 ready-made HyperFrames blocks and components installed as editable source - transitions, overlays, titles, kinetic type, captions, charts, map overlays, 3D, camera moves. Use it before hand-building any named visual (CRT scanlines, glitch, film grain, a chart); also for hyperframes add and catalog.
hyperframes-creative
heygen-com/hyperframes
Non-animation creative direction for HyperFrames videos. Use for design spec (frame.md / design.md) handling, palettes, typography, narration, beat planning, audio-reactive visuals, composition patterns, and brand / style decisions. For atomic motion patterns and scene blueprints, use hyperframes-animation.
ai-video-generation
101-skills/superpowers
Generate AI videos with Google Veo, Seedance 2.0, HappyHorse, Wan, Grok and 40+ models via inference.sh CLI. Models: Veo 3.1, Veo 3, Seedance 2.0, HappyHorse 1.0, Wan 2.5, Grok Imagine Video, OmniHuman, Fabric, HunyuanVideo. Capabilities: text-to-video, image-to-video, reference-to-video, video editing, lipsync, avatar animation, video upscaling, foley sound. Use for: social media videos, marketing content, explainer videos, product demos, AI avatars. Triggers: video generation, ai video, text to video, image to video, veo, animate image, video from image, ai animation, video generator, generate video, t2v, i2v, ai video maker, create video with ai, runway alternative, pika alternative, sora alternative, kling alternative, seedance, happyhorse
ai-image-generation
101-skills/superpowers
Generate AI images with GPT-Image-2, FLUX, Gemini, Grok, Seedream, Reve and 50+ models via inference.sh CLI. Models: GPT-Image-2, FLUX Dev LoRA, FLUX.2 Klein LoRA, Gemini 3 Pro Image, Grok Imagine, Seedream 4.5, Reve, ImagineArt. Capabilities: text-to-image, image-to-image, inpainting, LoRA, image editing, upscaling, text rendering. Use for: AI art, product mockups, concept art, social media graphics, marketing visuals, illustrations. Triggers: flux, image generation, ai image, text to image, stable diffusion, generate image, ai art, midjourney alternative, dall-e alternative, text2img, t2i, image generator, ai picture, create image with ai, generative ai, ai illustration, grok image, gemini image, gpt image, openai image, chatgpt image