agentleFS
orbyn-devorbyn-dev/orbyn-fullstack-skills/skills/social-preview/SKILL.md

social-preview

Skill4 starsChanged 10 days agoMITView on GitHub

Create and inspect local social post previews, announcement graphics and carousels for LinkedIn and other platforms. Use for feed mockups and exportable social visuals, not product UI changes or publishing.

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    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:

    Create reviewable visuals from a post or brief, for personal work as well as Orbyn announcements.

    line 8
  2. Distinctive:

    Use approved copy, supplied screenshots and brand assets.

    line 25

≈ “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

Social visuals and previews

Create reviewable visuals from a post or brief, for personal work as well as Orbyn announcements. Keep the editable source alongside requested exports. Prefer installed fonts/tools and supplied captures. Report missing export tools or current platform verification without inventing results. The local pinned design reference is optional offline background.

Choose the artifact

Decide whether the request needs a feed approximation, an attached visual, a product demonstration or a website link-preview card. If unspecified, use the supplied material to choose a useful local preview and state the choice.

Drafting or materially rewriting the post belongs to social-content; polishing prose belongs to unslop.

Use approved copy, supplied screenshots and brand assets. For Orbyn visuals, read the relevant stack/token owners and reference assets; do not invent a logo or treat the helper's default colors as official branding. Label conceptual UI and synthetic data as mockups. Keep actual product screenshots faithful; annotations must not imply behavior the captured product lacks.

Design the visual

Choose hierarchy, spacing and a visual explanation before adding decoration. Make each card legible at feed size. Show the product, diagram or concrete result when it explains more than a text card. For original illustration or image edits, use the available image-generation tool; typeset final exact copy separately. Never assume Codex-only tool names exist in Claude or install a provider implicitly.

Render the output

  • Quick text/card/carousel layout: use the dependency-free renderer and brief guide. It creates editable HTML with approximate wide/narrow feed views, full post text and ordered cards. Its example brief is a reusable input example, not a brand or personal voice profile. Extend or replace the layout for the actual brief rather than forcing every graphic into its starter template.
  • Product screenshot composition: use available browser/layout tools and supplied captures.
  • Requested video demo: define the actual demonstrated flow, use an available capture/export tool and inspect playback. A storyboard is not a video.
  • Requested subtitled video or engineering catch-up output: use the conditional video method; verify the actual encoded file and frame holds.
  • LinkedIn document carousel: export a PDF when requested, using an available PDF/presentation tool or browser print path. Read the relevant host skill when available. The bundled helper itself exports HTML and text only.

Inspect the rendered result

Open the result with an available browser or artifact viewer. Inspect the wide and narrow feed, every card, long words, line breaks, image crops and contrast. Check that sources and limitations survived the visual edit. Revise overflow and cramped copy rather than shrinking everything. Use useful alt text; keep decorative context out of it. Inspect exported PNG/PDF files separately at their actual dimensions, including every PDF page and its order.

Deliver honestly

Return the editable source, requested exports, copyable caption and alt text. Say which outputs were actually rendered and inspected. If the host lacks a renderer or exporter, deliver the source with the missing step explicitly marked; do not claim source inspection proves visual quality or upload readiness.

Feed chrome, line clamps, fonts and viewports are local approximations. Exact LinkedIn truncation, upload behavior and engagement are unverified until checked in the relevant platform. Layout dimensions are design choices; confirm current platform requirements when necessary.

Website Open Graph changes belong to the owning website task, and product flow/layout changes to orbyn-ui-design. Faux product UI for a page hero or feature card belongs to product-illustration. Do not create a site, publish, schedule or modify account settings as a side effect of preview generation.

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.