guides/content/content-repurposing/SKILL.md
Content atomization — turn one piece of content into many formats. Covers blog-to-thread, blog-to-carousel, podcast-to-blog, video-to-quotes, and more. Use for: content marketing, social media, multi-platform distribution, content strategy. Triggers: content repurposing, repurpose content, content atomization, content recycling, one to many content, multi platform content, cross post, adapt content, reformat content, blog to thread, blog to video, podcast to blog, content multiplication
npx skillsauth add 1nfsh-s3/skills content-repurposingInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Install the belt CLI skill:
npx skills add belt-sh/cli
Turn one piece of content into many formats via inference.sh CLI.
Requires inference.sh CLI (
belt). Install instructions
belt login
# Generate a quote card from a blog pull-quote
belt app run falai/flux-dev-lora --input '{
"prompt": "minimal quote card design, dark navy background, large white quotation marks, clean sans-serif typography space, modern professional design, social media post format",
"width": 1024,
"height": 1024
}'
One source piece can generate 10+ derivative assets:
┌──────────┐
│ LONG-FORM │ Blog post, podcast, video, whitepaper
│ SOURCE │
└─────┬─────┘
│
┌─────────┼─────────┐
▼ ▼ ▼
┌─────────┐ ┌──────┐ ┌──────────┐
│ MEDIUM │ │MEDIUM│ │ MEDIUM │ Newsletter, LinkedIn, email
│ FORMAT │ │FORMAT│ │ FORMAT │
└────┬────┘ └──┬───┘ └────┬─────┘
│ │ │
┌────┼────┐ │ ┌────┼────┐
▼ ▼ ▼ ▼ ▼ ▼ ▼
┌───┐┌───┐┌───┐┌───┐┌───┐┌───┐┌───┐ Tweets, quotes, audiograms,
│ ││ ││ ││ ││ ││ ││ │ short clips, infographic tiles
└───┘└───┘└───┘└───┘└───┘└───┘└───┘
Extract 5-8 key insights. One per tweet. Add hook.
| Element | Rule | |---------|------| | Hook tweet | Listicle, contrarian, or promise format | | Body tweets | One insight per tweet, 280 chars max | | Visual breaks | Add image every 3-4 tweets | | Final tweet | CTA + "RT the first tweet if useful" |
Adaptation:
# Generate a visual for the thread
belt app run falai/flux-dev-lora --input '{
"prompt": "clean infographic tile, single statistic 60% highlighted in large bold text, minimal dark background, data visualization style, professional",
"width": 1024,
"height": 1024
}'
# Post the thread
belt app run x/post-create --input '{
"text": "I analyzed 500 landing pages.\n\nHere are 7 patterns the top converters all share:\n\n🧵 Thread:"
}'
1 slide per section. 8-12 slides total.
| Slide | Content | |-------|---------| | 1 (Hook) | Bold claim or question from headline | | 2-9 (Content) | One key point per slide, large text, supporting visual | | 10 (Summary) | Recap the key takeaways | | 11 (CTA) | "Follow for more" / "Save this" / "Comment your thoughts" |
Specs: 1080x1080 (square) or 1080x1350 (4:5 for more space)
# Generate carousel slides
for i in {1..10}; do
belt app run falai/flux-dev-lora --input "{
\"prompt\": \"clean minimal presentation slide, dark gradient background, large text area, professional business design, slide $i of 10, consistent style\",
\"width\": 1024,
\"height\": 1024
}" --no-wait
done
3-line summary + "why it matters" + link.
## This Week's Feature: [Title]
[1-2 sentence summary of the key insight]
**Why it matters:** [1 sentence connecting to reader's work/life]
→ [Read the full post](link)
Problem + key insight + CTA. Under 60 seconds.
| Section | Duration | Content | |---------|----------|---------| | Hook | 3s | "Most people get [topic] wrong." | | Problem | 10s | State the common mistake | | Insight | 25s | Your key finding/advice | | Proof | 10s | One stat or example | | CTA | 5s | "Follow for more" / "Link in bio" |
# Generate voiceover
belt app run falai/dia-tts --input '{
"prompt": "[S1] Most landing pages make this mistake. They put the features above the fold instead of the outcome. Top converting pages show what life looks like AFTER using the product. Try it and watch your conversion rate climb."
}'
# Generate video
belt app run google/veo-3-1-fast --input '{
"prompt": "Screen recording style, scrolling through a well-designed landing page, clean modern UI, smooth scroll, professional website"
}'
Pull best quote. Generate audio. Add waveform visual.
# Generate audio of the key quote
belt app run falai/dia-tts --input '{
"prompt": "[S1] The number one mistake I see on landing pages... is putting features above the fold. The best pages show the outcome. Not what your product does, but what life looks like after."
}'
# 1. Transcribe the episode
belt app run <stt-app> --input '{
"audio": "episode-42.mp3"
}'
# 2. Edit transcript into blog format:
# - Remove filler words (um, uh, like, you know)
# - Add headers at topic changes
# - Break into paragraphs
# - Add intro and conclusion
# - Add links mentioned in the episode
3-5 best quotes with speaker attribution.
# Generate quote card backgrounds
belt app run falai/flux-dev-lora --input '{
"prompt": "minimal quote card background, subtle gradient from dark blue to black, large quotation mark watermark, clean modern design, social media square format",
"width": 1080,
"height": 1080
}'
Key moment, 3-5 seconds, under 5MB.
Best moments for GIFs:
# Extract the best 15-60 second segments for Reels/TikTok/Shorts
# Focus on self-contained moments that make sense without context:
# - A single tip or insight
# - A surprising stat reveal
# - A demonstration of one feature
# - A strong opinion or hot take
Never copy-paste across formats. Each platform has different:
| Platform | Attention Span | Tone | Format | |----------|---------------|------|--------| | Blog | 5-10 min | Thorough, detailed | Long paragraphs OK | | Twitter/X | 5-30 sec per tweet | Punchy, declarative | 280 chars, fragmented | | LinkedIn | 1-3 min | Professional, insightful | Short paragraphs, line breaks | | Newsletter | 5-7 min | Curated, personal | Sections with headers | | TikTok/Reels | 15-60 sec | Energetic, direct | Hook in 1 second | | Podcast | 20-60 min | Conversational, deep | Stories and tangents OK |
For each piece of long-form content, create:
| Mistake | Problem | Fix | |---------|---------|-----| | Copy-pasting between platforms | Feels lazy, wrong format | Rewrite for each platform's style | | Repurposing weak content | Amplifies mediocrity | Only repurpose your best pieces | | Same day posting everywhere | Audience overlap sees duplicates | Stagger across days/weeks | | Losing the core message | Derivative misses the point | Identify the ONE key insight first | | No visual adaptation | Text-only on visual platforms | Create platform-specific graphics | | Forgetting attribution | Plagiarizes yourself | Link back to the original |
npx skills add inference-sh/skills@ai-social-media-content
npx skills add inference-sh/skills@ai-image-generation
npx skills add inference-sh/skills@text-to-speech
npx skills add inference-sh/skills@twitter-automation
Browse all apps: belt app store
data-ai
Generate multi-person talking head podcast videos from scratch using AI — character creation, TTS, avatar animation, and video stitching. Use when the user wants to create a podcast, talking head video, or multi-speaker conversation video.
development
Build and deploy applications on inference.sh. Use when getting started, understanding the platform, creating apps, configuring resources, or needing an overview of inference.sh app development. Supports both Python and Node.js. Triggers: inference.sh app, belt app, inf.yml, inference.py, inference.js, deploy app, app development, build app, create app, GPU app, VRAM, app resources, app secrets, app integrations, multi-function app
data-ai
Generate multi-person talking head podcast videos from scratch using AI — character creation, TTS, avatar animation, and video stitching. Use when the user wants to create a podcast, talking head video, or multi-speaker conversation video.
tools
Generate videos with ByteDance Seedance 2.0 via inference.sh CLI. Unified model for text-to-video, image-to-video, and reference-to-video with synchronized audio, up to 1080p, 4-15s duration. Pro and Fast variants. Studio variants with private asset library for portrait consistency. Use for: social media videos, music videos, product demos, animated content, AI video with sound. Triggers: seedance, seedance 2, bytedance video, seedance t2v, seedance i2v, seedance r2v, video with audio, seedance 2.0, bytedance seedance, seedance studio