article-cover-image

Turns one title (or a full article) into a single 16:9 editorial cover — brand color #E63A46, MiSans-style typography, matte non-plastic finish.

When to use it

Title only

I have a headline or single-sentence topic and I want the skill to jump straight to a cover image — no long analysis first.

Title + article

I have a full draft ready and need one horizontal cover for a WeChat post, a video, a PPT title page, or a website banner.

Prompt only

I want the raw image prompt so I can run it in another model. The skill switches to text mode and does not call image generation.

Text-free

I will typeset the title myself. I just need the background artwork with a clean text-safe area.

Style hint

I do not want yet another dark blue tech look. I ask for editorial magazine, cinematic, or matte 3D-icon style and the skill routes to a different lane in its style pool.

Not for

• icons or black-line symbols → black-line-icon-style
• in-body article illustrations (not the cover) → post-illustration-images
• infographics, knowledge cards, comparison tables, process long-scrolls → the infographic workflow
• brand logo finalization → the logo design workflow
• pure copywriting or non-visual analysis → the writing workflow

What you get

By default the skill generates the image directly — no plan, no long preamble unless you explicitly say "prompt only", "no image", or "analyze first".

• Cover image: one 16:9 horizontal image, brand color #E63A46 clearly present, MiSans-style Chinese typography, matte low-reflection finish
• Title compression: a long headline gets compressed to a cover title of at most 12 Chinese characters (e.g. "Why the default workflow belongs in AGENTS.md, not a Skill" → "AGENTS.md 优先")
• Auto style routing: picks one of 8 lanes — minimal magazine, modern product UI, cinematic metaphor, young playful illustration, editorial business, matte 3D icon, abstract art, or futuristic tech — based on the topic keywords
• Text-mode output: when you ask for the prompt only, you get the main Prompt plus a Negative Prompt, ready to paste elsewhere
• Never: generates infographics, VS comparison layouts, multiple main subjects, glossy plastic or toy 3D looks, subtitles, or long body copy

Prerequisites & boundaries

Prerequisites

An image-generation runtime — without it, only "prompt only" mode works. At least one title, or a topic recoverable from the article body.

Adjacent skills

Task	Route to
icon or black-line symbol	black-line-icon-style
in-body article illustration	post-illustration-images
infographic / knowledge card / process diagram	infographic workflow
formal brand logo	logo design workflow

Not accepted

• No title given and none extractable from the body — the skill stops and asks
• Article copywriting or generic writing prompts
• Non-visual analysis or general prompt engineering

Subtle boundaries

• "I want an illustration" — a single horizontal cover triggers this skill; multiple in-body illustrations route to post-illustration-images
• "I want an infographic or knowledge card" — does not trigger; this skill is deliberately built to prevent infographic drift
• "Prompt only" — a cover-image target triggers this skill in text mode; a writing or general prompt target does not