agentleFS
pabloirracionalpabloirracional/claude-skills/skills/designing-carousels/SKILL.md

designing-carousels

Skill0 starsChanged 4 months agoMITView on GitHub

Builds multi-card social media carousels (Instagram/LinkedIn-style) with a consistent brand system, a clear per-card content structure, and export-ready output. Make sure to use this skill whenever the user wants a carousel, multi-slide social post, wants to turn a blog post/article/thread into shareable cards, or is working in Claude Design (claude.ai/design) on a set of slides/cards - even if they just say "carrossel" or "cards pro Instagram/LinkedIn".

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Never generate cards before this is set - skipping it produces generic, AI-looking output.

    line 26
  2. Distinctive:

    If the user has past carousels or brand assets, ask for a reference image instead of asking them to describe it - matching a reference is …

    line 32
  3. Distinctive:

    Write the text for every card as a numbered outline before generating any visuals.

    line 45
  4. Distinctive:

    Keep every card visually consistent: same color system, same type hierarchy, same margin, same logo position.

    line 49
  5. Distinctive:

    Default aspect ratios unless the user specifies otherwise:

    line 51
  6. Distinctive:

    If the user mentions Claude Design (claude.ai/design), the workflow above maps directly onto it: brand system setup, card-by-card …

    line 65
  7. Distinctive:

    If no brand system exists yet, offer to help define one from an existing post, website, or logo before starting the carousel - it pays off …

    line 66

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Designing carousels

Turns a piece of content into a consistent, on-brand multi-card carousel - the kind used on Instagram and LinkedIn.

Workflow

Copy this checklist and track progress:

Progress:
- [ ] Step 1: Confirm the brand system
- [ ] Step 2: Define card count and role
- [ ] Step 3: Outline content per card
- [ ] Step 4: Generate cards
- [ ] Step 5: Review and iterate
- [ ] Step 6: Export

Step 1: Confirm the brand system

Never generate cards before this is set - skipping it produces generic, AI-looking output. Confirm or ask for:

  • Color palette (primary, secondary, background, text)
  • Typography (heading font, body font - or a reference style if the user doesn't know font names)
  • Logo usage and placement
  • Spacing/margin convention (how much breathing room around content)

If the user has past carousels or brand assets, ask for a reference image instead of asking them to describe it - matching a reference is far more reliable than a verbal brief.

Step 2: Define card count and role

Most carousels follow this structure - adapt count to content depth, but keep the roles:

  1. Hook card - the stop-scroll headline. No logo-heavy branding here, just the promise.
  2. Content cards (3-8) - one idea per card. Never cram two points onto one card.
  3. Summary/recap card (optional) - condenses the key takeaways for skimmers.
  4. CTA card - what to do next (follow, comment, link in bio, save).

Step 3: Outline content per card

Write the text for every card as a numbered outline before generating any visuals. This catches pacing problems (too much text on one card, a weak hook, a missing CTA) while it's still cheap to fix.

Step 4: Generate cards

Keep every card visually consistent: same color system, same type hierarchy, same margin, same logo position. Only the content changes card to card - if the layout style shifts between cards, the set reads as unfinished.

Default aspect ratios unless the user specifies otherwise:

  • Instagram: 4:5 portrait (1080×1350px) - maximizes feed real estate
  • LinkedIn: 1:1 square (1080×1080px) - safest across LinkedIn's crop behavior

Step 5: Review and iterate

Expect 2-3 rounds before it's on-brand. Common fixes: text too small on mobile preview, inconsistent margins between cards, a reference style not fully applied. If a reference image is available, ask the tool to re-benchmark against it explicitly rather than describing the drift in words.

Step 6: Export

Export each card as an individual PNG at the target dimensions - never as one combined image. Confirm the background/canvas is excluded and only the card itself is exported.

Notes

  • If the user mentions Claude Design (claude.ai/design), the workflow above maps directly onto it: brand system setup, card-by-card generation, then PNG export per card.
  • If no brand system exists yet, offer to help define one from an existing post, website, or logo before starting the carousel - it pays off across every future carousel, not just this one.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.