Skip to content

rampstackco/claude-skills / skills/content-repurposing

content-repurposing

Verified

Cross-format content adaptation. Turning one substantial piece into many derivative formats (blog series, email sequences, social posts, webinars, podcasts, video shorts) without losing the original's value or producing AI-slop variants. The discipline of adaptation per medium rather than mass-blast distribution. Triggers on content repurposing, content adaptation, cross-format content, content atomization, content multiplication, content distribution across formats, source-piece-to-derivative, video shorts from blog, email from whitepaper, podcast from article, blog series from research. Also triggers when a flagship piece is shipping but the team has not planned how to extend it across formats, when repurposing is happening but the derivatives feel mass-produced, or when AI-assisted repurposing is producing slop variants of strong source pieces.

references/per-format-constraints.md

View on GitHub
# Per-format constraints

What each medium demands and forbids. Length norms, conventions, hooks, and the underlying observation: derivatives that ignore the medium's constraints fail the medium even when the source content was strong.

Each format below has specific reader expectations. Pieces that meet expectations earn engagement; pieces that violate expectations get scrolled past or ignored regardless of content quality. Repurposing capacity goes to waste when derivatives fail the format-specific tax.

---

## Email

**Length norms.** 200-800 words per email for most B2B and consumer audiences. Newsletter-style emails can run longer (1,500-3,000 words) for engaged audiences but shorter is generally safer.

**What email demands.**

- A subject line that earns the open. Most engagement happens or fails at the subject-line level.
- A preheader that supports the subject line. The preheader is the second-most-read element of the email and most templates underuse it.
- Plain prose paragraphs that render reliably. Many email clients break complex formatting; tables, columns, and rich layouts can fail.
- A single clear CTA. Multi-CTA emails dilute; readers process one action at a time.
- Mobile-first design. 60-70% of email opens are on mobile; readability on small screens is essential.

**What email forbids.**

- Image-only emails. Many clients block images by default; image-only emails arrive blank.
- Link-laden text dumps. More than 3-4 links in body text dilutes attention.
- Multi-CTA layouts that compete for the click.
- Subject lines that announce the topic without earning the open ("Newsletter for week of X").

**Adaptation discipline.** Email derivatives need new ledes (the source's lede does not transfer); spoken-direct voice (more "you" usage; shorter sentences than blog voice); a single CTA per email even when the source piece had multiple potential actions.

---

## LinkedIn (long-form text post)

**Length norms.** 800-2,500 characters typical; long-form posts up to ~3,000 characters. The character count matters because the platform truncates with "see more"; the truncation point is the hook.

**What LinkedIn demands.**

- A first line that earns the "see more" click. The first 1-3 lines determine whether readers expand the post.
- Clear structure: short paragraphs, white space, occasional emphasis. Walls of text get scrolled past.
- A specific point or argument. Narrative-style posts that take time to land work; vague observations do not.
- Engagement hooks: questions, polarizing claims, specific examples that invite comment.
- Original perspective. Repurposed content marked clearly as such; uncredited recycling reads as low effort.

**What LinkedIn forbids.**

- Walls of text without paragraph breaks.
- Generic motivational content; the platform has decreasing tolerance for this.
- Posts that read as ads with no substantive content.
- Heavy use of emoji decoration without purpose.

**Adaptation discipline.** Source content needs to be restructured for LinkedIn rhythm: short opening hook, body with paragraph breaks every 2-3 sentences, specific examples or stakes, possibly a question at the close that invites comment.

---

## X/Twitter (single post and threads)

**Length norms.** Single post: 280 characters (or up to 25,000 for premium accounts; conventions vary). Thread: 5-15 posts typical for substantive threads.

**What X demands (single post).**

- Tight phrasing. Each character matters; specific framings that fit the format earn engagement.
- Stand-alone meaning. Tweets that depend on context outside the tweet often fail.
- Specific points or claims. Vague observations underperform.

**What X demands (threads).**

- A first post that earns the next click. Threads are read sequentially; the first post must hook.
- One coherent argument across the thread. Threads that wander get abandoned mid-way.
- Reasonable length. Threads longer than 15-20 posts often lose readers; very long threads may be better as a blog post linked from a shorter thread.

**What X forbids.**

- Threads that read as a blog post split into 280-character chunks. The platform's conversational rhythm is different.
- Hashtags-as-content. Heavy hashtag usage signals SEO chasing rather than communication.

**Adaptation discipline.** Source content for threads needs to be re-paced: each post a complete thought, the thread building, the conclusion specific.

---

## Long-form video (YouTube, in-depth content)

**Length norms.** 8-30 minutes typical for substantive content. Some categories support longer (45-90 minutes for deep technical content); the algorithm rewards retention over duration.

**What long-form video demands.**

- Pacing for spoken delivery. Print-style sentences read aloud sound stilted; video scripts need spoken voice.
- Visual variation. Talking-head video alone for 30+ minutes underperforms; cuts, b-roll, on-screen text, illustrations carry attention.
- Audio quality. Bad audio kills retention faster than bad video; audio investment usually returns more than video investment.
- A specific topic. The video's title and thumbnail need to communicate scope; vague titles underperform in algorithmic recommendation.

**What long-form video forbids.**

- Slideshows of text read aloud. Mass-blast pattern in video form.
- No editing. Filler words, long pauses, and unedited rambling reduce retention.
- Talking head only with no visual variation across long durations.

**Adaptation discipline.** Source content for video needs scripting in spoken voice (or talking-points for unscripted delivery), visual planning per segment, and an opening that earns the next 30 seconds of viewing.

---

## Short-form video (TikTok, Reels, Shorts)

**Length norms.** 60-90 seconds typical; 15-30 seconds for some platforms. Algorithmic recommendation typically rewards retention from second 1.

**What short-form demands.**

- Hook in seconds 1-2. Viewers scroll past the opening if it does not earn the next 5 seconds.
- Captions. Many viewers watch without sound; captions carry the message.
- A single specific point. Short-form does not support multi-point delivery.
- Visual presence. Motion, text-on-screen, or visual variation within the short. Static talking heads underperform.

**What short-form forbids.**

- Content that takes 10+ seconds to land. Viewers leave before the point.
- Long-form video clipped without re-pacing. The pacing assumed in the source piece does not match short-form expectations.
- Captions that lag the audio. Mistimed captions distract.

**Adaptation discipline.** Source content for shorts needs extreme compression: one moment, one point, one short. Each short is a standalone unit.

---

## Podcast

**Length norms.** 20-90 minutes typical for substantive content. Daily-news-style podcasts can be shorter (8-15 minutes); deep-interview podcasts can run 90-180 minutes.

**What podcast demands.**

- Spoken-language sentence structure. Print-style sentences read aloud sound stiff.
- Conversational pacing. Examples, asides, and natural speech earn engagement; tightly-argued print sections feel cold.
- Audio quality. Listening is intimate; bad audio breaks the connection. Better audio than video matters here.
- Show notes that do the print work the audio cannot. Links, structured summary, citations, references.

**What podcast forbids.**

- Reading articles aloud without adaptation.
- Scripts that sound scripted (over-rehearsed delivery, no conversational rhythm).
- Episodes without show notes or with thin show notes.
- Inconsistent episode length within a series; audiences develop expectations.

**Adaptation discipline.** Source content for podcasts becomes the show's spine; the host adds context, examples, and conversational elaboration. Show notes are a distinct derivative format with their own constraints.

---

## Webinar

**Length norms.** 30-60 minutes typical. Longer webinars (90+ minutes) require very high engagement design or strong specific draws (like premiere announcements).

**What webinars demand.**

- A live element. Q&A, polls, interactive moments. Static webinars are videos; interactive webinars earn the format.
- Slide design that supports rather than competes with delivery. Text-heavy slides distract; visual slides with the speaker carrying the content work.
- Pacing for live audience attention. Audience attention dips in the middle; engagement design needs to address the dip.
- Recording strategy. Most webinars become video assets after the live session; the recording extends value indefinitely.

**What webinars forbid.**

- Reading slides aloud. The audience can read; the speaker should add the value the slides cannot carry.
- No engagement design. Audiences disengage in pure one-way webinars.
- Webinars that try to cover too much in 30 minutes. Cure: narrow the topic to what 30 minutes can deliver.

**Adaptation discipline.** Source content for webinars typically takes the source's 3-5 most load-bearing sections as the spine, plus interactive moments, plus Q&A.

---

## AI-search snippet (for citation by AI engines)

**Length norms.** 40-80 words per snippet for FAQ-style extractions; 100-200 words for paragraph-style extractions.

**What AI-search snippets demand.**

- Standalone meaning. The snippet works without the surrounding context; AI engines extract paragraphs, not full pieces.
- Named-source citations within the snippet. AI engines weight named-source claims; "according to a 2025 Forrester study" performs better than "studies show."
- Specific framings. Vague phrasings rarely get cited; specific phrasings do.
- Schema markup where applicable. FAQPage and similar schemas signal the format.

**What AI-search snippets forbid.**

- Floating claims without sources.
- Pronoun references that depend on outside context ("This is why..." with no clear referent within the snippet).
- Hedged language on claims that can be specific.

**Adaptation discipline.** Source content for AI snippets needs to be extractable as standalone paragraphs with internal citations and specific framings. Often the source piece needs additions specifically designed for snippet citation.

---

## Methodology-level choices that stay in the public skill

The format constraints by medium, the length norms, the demand-and-forbid lists, the adaptation discipline per format. The underlying observation: medium constraints are non-negotiable.

## Implementation choices that stay internal

Specific platform automation. Specific design templates per format. Specific captioning workflows. Specific schema markup patterns. Specific A/B testing of format-specific elements (subject lines, hooks, opening lines). The team's own conventions on length within the bands. These vary by team and platform mix.
add-product-e2e-evalpaperclipai/paperclipAdd or extend a Paperclip full-stack runner E2E workflow, fixture, matcher, or report evidence path for local or Daytona execution.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
diagnose-why-work-stoppedpaperclipai/paperclipCompatibility entry point for failed, stalled, looping, or over-recovered Paperclip work. Use fix-failed-paperclip-run for the full diagnosis, regression, repair, real verification, and green-PR workflow; honor an explicitly requested diagnosis-only scope.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
fix-failed-paperclip-runpaperclipai/paperclipDiagnose and fix failed, stuck, slow, or incorrectly blocked Paperclip tasks and runs. Use with a task/run URL or failure report to trace the root cause, prove a regression red then green, check second-order effects, verify the real workflow, and deliver a PR with passing checks.Stars: 115kUpdated Claude CodeCodexVerifiedInstall