Hubbestaiskill.com

Home / All Skills / Content & Media Creation / canghe-infographic

Content & Media Creation · freestylefly/canghe-skills

canghe-infographic

Generates professional infographics with 20 layout types and 17 visual styles. Analyzes content, recommends layout×style combinations, and generates publication-ready infographics. Use when user asks to create "infographic", "信息图", "visual summary", or "可视化".

L1 Basic · Source registration onlyV1 Commit-PinnedDual-pass Audit

Author freestyleflyGitHub freestylefly/canghe-skills ↗License MIT（仓库根 LICENSE）Commit dd0bf35595

Control levels describe skill capabilities and suggested control measures; they do not represent quality or safety certifications, nor do they dictate usability. Usage decisions rest entirely with users and administrators. Different agent hosts apply varying constraints; do not assume hosts provide automatic protection. Third-party skill copyrights belong to their original authors. Use at your own risk.

1How It Works & Mechanism

The core is the "two-dimensional free combination" model: arbitrary combinations of information structure layout (20 types) × visual style (17 types), defaulting to bento-grid + craft-handmade.

skills/canghe-infographic/SKILL.md

Two dimensions: **layout** (information structure) × **style** (visual aesthetics). Freely combine any layout with any style.

What is happening here: Orthogonally separating "how to arrange information" from "what it looks like", serving as the organizational backbone of the entire workflow.

7-step workflow: 1 Read EXTEND.md preferences → 2 Analyze content and write analysis.md → 3 Generate structured-content.md → 4 Recommend 3-5 combinations → 5 Single confirmation → 6 Generate prompt → 7 Invoke image skill to generate images.

skills/canghe-infographic/SKILL.md

1. Select available image generation skill (ask user if multiple)

What is happening here: Step 7 does not have built-in image generation capabilities, but instead directs the agent to use an image skill installed on the host.

Content fidelity discipline: Source data must be preserved verbatim, must not be summarized or rewritten, and no new information may be added.

skills/canghe-infographic/SKILL.md

Preserve all source data **verbatim**—no summarization or rephrasing

What is happening here: Learning objectives are determined first, followed by visual communication organization (headings/labels/visual elements), but the data itself cannot be altered.

EXTEND.md preferences are probed in a two-tier order (project-level → user-level) using Bash test -f; if found, they are read and parsed; if not found, AskUserQuestion is used for initial configuration guidance.

skills/canghe-infographic/SKILL.md

test -f .canghe-skills/canghe-infographic/EXTEND.md && echo "project"

What is happening here: This is purely an existence check without modifying files; the outcome determines which preference to read or whether to trigger initial setup.

Backup discipline: If source.md / analysis.md / prompts/infographic.md / infographic.png already exist, rename them by appending -backup-YYYYMMDD-HHMMSS before writing to avoid silent overwrites.

skills/canghe-infographic/SKILL.md

**Backup rule**: If `source.md` exists, rename to `source-backup-YYYYMMDD-HHMMSS.md`

What is happening here: Defined identical-name backup rules for four types of artifacts to prevent overwrites.

The image prompt is assembled across four layers: corresponding layout definition + corresponding style definition + base-prompt template + structured content from Step 2, with all text in the confirmed language.

skills/canghe-infographic/SKILL.md

1. Layout definition from `references/layouts/<layout>.md`

What is happening here: Prompts are not created freely, but assembled systematically from templates (fetching one reference file each for layout and style).

Image generation is delegated to an external image skill: provide the prompt file and output path, automatically retry once upon failure, and output artifacts to the infographic/{topic-slug}/ directory.

skills/canghe-infographic/SKILL.md

3. Call with prompt file and output path

What is happening here: This skill only produces prompts; actual image generation relies on the selected image generation skill.

Each of the 20 layouts and 17 styles has its own reference file; the main text contains only indices and "applicable scenarios", reading single files only when needed (progressive disclosure, saving context).

skills/canghe-infographic/SKILL.md

Full definitions: `references/layouts/<layout>.md`

What is happening here: SKILL.md lists only quick-reference tables for layouts/styles, fetching detailed definitions from references/ on demand.

2Core Capabilities

0120 information layouts (timeline, comparison, hierarchy, funnel, iceberg, journey, etc.)

0217 visual styles (craft-handmade, clay, kawaii, cyber-neon, blackboard, etc.)

03Automatically recommends 3-5 layout×style combinations based on content

04Single AskUserQuestion to consolidate confirmation of combination, aspect ratio, and language

05Structured content templates (titles, learning objectives, sections, verbatim data points)

06Output directory and naming discipline (topic-slug, timestamp on conflict, backup renaming)

07Aspect ratios (16:9 / 9:16 / 1:1) and multilingual support (--lang en/zh/ja…)

08EXTEND.md custom preferences (preferred layout/style, default ratio, custom styles, language)

3External Dependencies

Kind	Dependency / Endpoint
cli	bash（SKILL.md 用 test -f 检查 EXTEND.md 是否存在，无副作用）

4Recommended Governance & Evidence

Suggested Controls for Safe Operation

For Agent Runtime

Can be auto-loaded

For Administrators

Register source and version

Evidence & Triggers (File Paths & Excerpts)

SKILL.md: Pure prompt instructions and rule text, without script execution or external impact

Audit Notes: Green: There are no scripts or executable files within this skill directory (all 39 files are Markdown), no network endpoints, and no API key/credential reading; the only shell operation is a side-effect-free `test -f` existence check in SKILL.md. File writing of artifacts and image rendering actions are performed by the host agent or a separately installed image skill, which does not affect this skill's tier under the "assessment target = skill directory itself" criterion. Step 6 explicitly delegates execution to an external image generation skill (cross-skill delegation), which has been documented under risks and does not elevate the risk tier.

5Second-Pass Verification

[ok] 是否含脚本或可执行文件 — §1 清单 41 个文件全部为 .md（SKILL.md + analysis-framework + base-prompt + structured-content-template + 20 layouts + 17 styles），无 .py/.sh/.js；SKILL.md 唯一 shell 是 test -f 存在性检查，无副作用。
[discrepancy] SKILL.md 引用的 references 是否都存在 — SKILL.md Step 1.1 与 Extension 段引用 `references/config/first-time-setup.md` 和 `references/config/preferences-schema.md`，但 §1 清单中 references/ 下只有 analysis-framework.md、base-prompt.md、structured-content-template.md、layouts/、styles/，无 config/ 目录——这两处引用为悬空，首次配置流程与偏好 schema 实际缺失。
[ok] 布局/风格数量口径 — SKILL.md 宣称 20 layout / 17 style；§1 清单 references/layouts/ 恰 20 个 md、references/styles/ 恰 17 个 md，吻合。
[ok] 是否有外发或凭证读取 — 全文无网络端点、无 API key/环境变量读取；$HOME 仅出现在 EXTEND.md 存在性检查与路径展示处。
[ok] 出图由谁执行 — Step 6 明示 Select available image generation skill（多个则询问用户），由宿主另装 skill 出图，本 skill 不自带任何模型调用。
[ok] 备份与覆盖行为 — 对 source.md / analysis.md / prompts/infographic.md / infographic.png 均定义先改名加 -backup-YYYYMMDD-HHMMSS 的备份规则，非静默覆盖。
[ok] quote 逐条复核 — 本报告全部 evidence.quote 以子串方式回 bundle 原文比对，无编造。

6Verdict & Scenarios

Orthogonally separates "information structure" and "visual style" into 20×17 combinations, accompanied by indices and recommendation tables, ensuring controllable and predictable outputs.

Strict content fidelity discipline (verbatim retention, no additions) and explicit interaction discipline (single confirmation) significantly reduce divergence.

Pure prompt + progressive disclosure design with zero dependencies and zero credentials, offering a minimal security footprint that is easy to audit and portable.

Best for: Creators and content teams converting articles or materials into infographics, long-form graphics, or visual summaries; also suitable as a prompt orchestration template for secondary customization.

Not for: Scenarios expecting the skill to include a built-in image model or one-click offline generation; scenarios requiring scripted batch processing (this skill contains no scripts and relies on manual interactive confirmation).

Install Agent-Ready

① Verified Mirror 2026-10-10

Option A · Direct DownloadDownload canghe-infographic.tar.gz

sha256: 8a13ef43ede1ecb6…

Option B · Copy Guide for Agent

Agents automatically download and verify mirrors from this JSON guide.

② Upstream GitHub

GitHub Repo ↗

Pinned to commit dd0bf35595.