Baoyu Image Cards
Turns an article into a 1-10 card infographic series: 12 styles, 8 layouts, 3 palettes
Test report
- Verdict
- Works with setup
- Score
- Tested
- Jul 21, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · ec4246b
Cloned the repo and installed into a throwaway HOME (export HOME=$(mktemp -d)); SKILL.md landed at ~/.claude/skills/baoyu-image-cards/SKILL.md, 28 files, all markdown, zero scripts and zero curl/wget/base64/eval hits across the tree. Verified the frontmatter's claims: 12 files in references/presets, 3 in references/palettes, 8 layout rows in elements/canvas.md, and four spot-fetched references (style-presets.md, workflows/prompt-assembly.md, config/first-time-setup.md, elements/canvas.md) all returned HTTP 200. On a 150-word "5 mistakes that make your AWS bill explode" source I wrote a baseline card plan plus cover prompt first, then followed the body strictly (Step 0 found no EXTEND.md, Step 1 analysis.md, Step 2 Path A, Step 3 prompt file): the baseline cover prompt was 10 lines at 4:5 with "muted blue and orange palette on a light background" and no safe zones, while the skill-assembled prompts/01-cover-aws-bill-mistakes.md is 88 lines carrying the bold preset's exact hexes (#E53E3E/#DD6B20/#F6E05E on #000000, accent #F7FF00), 3:4 at 1242x1660, a 60-70% whitespace sparse spec, and explicit XHS safe zones (bottom 10% title bar, top-right like/share, bottom-right watermark) with no baseline equivalent. I could NOT execute Step 3.2 (generate image 1) — no raster backend exists in this runtime (no `imagegen`, no `baoyu-imagine`, no native image_generate) and the skill hard-forbids SVG/HTML substitutes — so no PNG was produced and the image-1 anchor chain is untested; hence outputMeasured=false and verdict setup.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 5/5
- Output vs. baseline 5/10
- Docs & honesty 4/5
What Baoyu Image Cards does
Breaks a source article or note into a 1-10 image-card series for Xiaohongshu, WeChat and other social feeds, writing an analysis file, an outline with per-card swipe hooks, and one full prompt file per card before any rendering. It picks style, layout and palette from content-type signals, then hands the assembled prompts to whatever raster image backend is installed, chaining card 1 as the reference image for visual consistency. Triggers on requests for 小红书图片, 微信图文, image cards, 图片卡片, or a social-media infographic card series.
How to install Baoyu Image Cards
git clone --depth 1 https://github.com/guanyang/open-agent-hub.git /tmp/baoyu-image-cards-src
mkdir -p ~/.claude/skills
cp -R /tmp/baoyu-image-cards-src/skills/baoyu-image-cards ~/.claude/skills/baoyu-image-cards
# 28 markdown files, no scripts, no runtime deps.
# REQUIRED EXTRA: the skill renders nothing itself. It resolves a raster image backend at
# Step 3 and explicitly forbids SVG/HTML/canvas substitutes. Install one of:
# - Codex CLI's built-in `imagegen` skill (auto-detected, highest priority), or
# - the `baoyu-imagine` skill, or another runtime-native image tool.
# With none installed the skill stops at Step 3 and asks how to proceed.
# Optional preferences file (style, layout, palette, watermark, backend, batch size):
# ~/.baoyu-skills/baoyu-image-cards/EXTEND.md (schema: references/config/preferences-schema.md)
# If absent, the first interactive run walks through first-time setup; `--yes` skips it.
# Usage: "turn this article into 小红书 image cards", or flags such as
# --preset warning --palette macaron --batch-size 4 --yes
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Baoyu Image Cards
-
/baoyu-image-cardsTurns an article into a 1-10 card infographic series: 12 styles, 8 layouts, 3 palettes
It also activates on plain-language prompts like these:
-
Make a Xiaohongshu image card series for this post -
Create infographic cards summarizing this article -
Generate a WeChat image-text card set on this topic
Frequently asked questions
- Is the Baoyu Image Cards skill free?
- Yes. The skill itself is free from guanyang/open-agent-hub. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Baoyu Image Cards work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 21, 2026. Verdict: Works with setup. Cloned the repo and installed into a throwaway HOME (export HOME=$(mktemp -d)); SKILL.md landed at ~/.claude/skills/baoyu-image-cards/SKILL.md, 28 files, all markdown, zero scripts and zero curl/wget/base64/eval hits across the tree. Verified the frontmatter's claims: 12 files in references/presets, 3 in references/palettes, 8 layout rows in elements/canvas.md, and four spot-fetched references (style-presets.md, workflows/prompt-assembly.md, config/first-time-setup.md, elements/canvas.md) all returned HTTP 200. On a 150-word "5 mistakes that make your AWS bill explode" source I wrote a baseline card plan plus cover prompt first, then followed the body strictly (Step 0 found no EXTEND.md, Step 1 analysis.md, Step 2 Path A, Step 3 prompt file): the baseline cover prompt was 10 lines at 4:5 with "muted blue and orange palette on a light background" and no safe zones, while the skill-assembled prompts/01-cover-aws-bill-mistakes.md is 88 lines carrying the bold preset's exact hexes (#E53E3E/#DD6B20/#F6E05E on #000000, accent #F7FF00), 3:4 at 1242x1660, a 60-70% whitespace sparse spec, and explicit XHS safe zones (bottom 10% title bar, top-right like/share, bottom-right watermark) with no baseline equivalent. I could NOT execute Step 3.2 (generate image 1) — no raster backend exists in this runtime (no `imagegen`, no `baoyu-imagine`, no native image_generate) and the skill hard-forbids SVG/HTML substitutes — so no PNG was produced and the image-1 anchor chain is untested; hence outputMeasured=false and verdict setup.
- What is the Baoyu Image Cards SkillProof Score?
- 7.6/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 5/10, docs & honesty 4/5.
- How do I install Baoyu Image Cards?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Baoyu Image Cards with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.