agentleFS
Marky-TeamMarky-Team/marky-skills/skills/create-post-carousel/SKILL.md

create-post-carousel

Skill5 starsChanged 4 months agoApache-2.0View on GitHub

Create a multi-slide carousel post — Instagram carousel, LinkedIn document post, or TikTok photo-mode — through the Marky API: pick a narrative framework, write the slide-by-slide copy, render every slide from one shared branded HTML template, and attach the slides in order. Use when the user says "make a carousel", "slide post", "LinkedIn document post", "turn this list into slides", or "swipeable post". For a single designed image, see create-post-graphic (this skill uses its render mechanics); for the caption and scheduling, see schedule-posts.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read references/carousel-frameworks.md and pick by what the content IS, using its table:

    line 24
  2. Distinctive:

    Don't force it: list content in a rant structure (or vice versa) is the #1 reason carousels flop.

    line 35
  3. Distinctive:

    Write ALL slides as text before touching HTML, following the framework's slide-by-slide table.

    line 41
  4. Distinctive:

    Run the slide copy and the caption through the humanize skill's pattern check before rendering — fixing text is free, re-rendering isn't.

    line 51
  5. Very specific:

    Use create-post-graphic's Stage 1 (brand pull), Stage 3 (HTML authoring), and Stage 4 (headless render + LOOK at it) with one …

    line 56

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Carousel Posts

A carousel is not a blog post chopped into squares — it's a swipe-through where every slide has two jobs: deliver one idea, and earn the next swipe. This skill owns the narrative structure and slide copy; the rendering mechanics (brand pull, HTML authoring, headless render, upload) come from the create-post-graphic skill — read it first.

Stage 1 — Pick the framework

Read references/carousel-frameworks.md and pick by what the content IS, using its table:

ContentFramework
List of resources/tools/tipsA: Value-Stack
A personal result with a system behind itB: Problem-Proof
Several named techniques on one themeC: Hack List
A strong opinion about a common practiceD: Rant Callout
A product or workflow you can showE: Demo Walkthrough

Don't force it: list content in a rant structure (or vice versa) is the #1 reason carousels flop. If nothing fits, it probably wants to be a single graphic — hand off to create-post-graphic.

Stage 2 — Write the slide copy

Write ALL slides as text before touching HTML, following the framework's slide-by-slide table. Rules from the reference that are non-negotiable:

  • Slide 1 must work as a standalone feed post (it competes alone as the thumbnail).
  • One idea per slide; if a slide needs two sentences of setup, split it.
  • Any count promised on the cover is exactly delivered — no filler slides.
  • Claims and numbers must be real and this business's own.
  • Final slide: ONE call to action, not three.

Run the slide copy and the caption through the humanize skill's pattern check before rendering — fixing text is free, re-rendering isn't.

Stage 3 — Render the slides

Use create-post-graphic's Stage 1 (brand pull), Stage 3 (HTML authoring), and Stage 4 (headless render + LOOK at it) with one carousel-specific rule: one HTML template, N content swaps. Author a single slide.html with the brand tokens and layout, then render each slide by swapping the copy (a small build loop or one file per slide from the same skeleton). Same type scale, same palette, same layout on every interior slide — variety between slides reads as clutter.

Dimensions: 1080x1350 (4:5) for Instagram and LinkedIn; check the marky-api platform rules for the max slide count per platform before authoring.

Stage 4 — Assemble and attach

  • Upload the slide PNGs in order (slide 1 first — it's the cover) and create the post with all of them in media_urls. Marky publishes them as a carousel on platforms that support it.
  • LinkedIn document posts: LinkedIn expects a PDF; Marky handles carousel delivery per platform — do NOT pre-convert unless the platform rules say so.
  • The caption is a second hook, not a repeat of slide 1 — write it as its own post text.
  • Tag metadata: format: carousel, hook: <framework letter>, plus topic — review-performance groups engagement by these tags, so framework performance becomes measurable per audience.

Stage 5 — Approve, then schedule

Same hard gate as every creation skill: show the rendered slides (review board in pick mode for multiple candidates), get explicit approval, then schedule via the schedule-posts skill. Judge results later on saves and completion, not likes (see "Measuring what worked" in the reference).

Feedback

Friction or ideas? Send via the submit_feedback MCP tool (see marky-api "Marky wants your feedback").

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.