Agent skill

Gimi Illustration

by GiMi-Xiaomi in GiMi-Xiaomi/gimi-illustration-skill

Install Gimi Illustration

skills CLI

$ npx skills add GiMi-Xiaomi/gimi-illustration-skill --skill gimi-illustration -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI

$ gh skill install GiMi-Xiaomi/gimi-illustration-skill gimi-illustration --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Claude Code skills documentation · loads skills from .claude/skills/

Paste into Claude Code

Install the "gimi-illustration" agent skill from https://github.com/GiMi-Xiaomi/gimi-illustration-skill/tree/main into .claude/skills/gimi-illustration/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "gimi-illustration", then confirm the skill loads.

Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add GiMi-Xiaomi/gimi-illustration-skill --skill gimi-illustration -a codex

Project install goes to .agents/skills/; add -g for ~/.codex/skills/.

GitHub CLI

$ gh skill install GiMi-Xiaomi/gimi-illustration-skill gimi-illustration --agent codex

Project scope by default (.agents/skills/); add --scope user for a personal install.

Codex skills documentation · loads skills from .agents/skills/

Paste into Codex

Install the "gimi-illustration" agent skill from https://github.com/GiMi-Xiaomi/gimi-illustration-skill/tree/main into .agents/skills/gimi-illustration/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "gimi-illustration", then confirm the skill loads.

Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add GiMi-Xiaomi/gimi-illustration-skill --skill gimi-illustration -a cursor

Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.

GitHub CLI

$ gh skill install GiMi-Xiaomi/gimi-illustration-skill gimi-illustration --agent cursor

Project scope by default (.agents/skills/); add --scope user for a personal install.

Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/

Paste into Cursor

Install the "gimi-illustration" agent skill from https://github.com/GiMi-Xiaomi/gimi-illustration-skill/tree/main into .cursor/skills/gimi-illustration/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "gimi-illustration", then confirm the skill loads.

Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add GiMi-Xiaomi/gimi-illustration-skill --skill gimi-illustration -a gemini-cli

Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.

GitHub CLI

$ gh skill install GiMi-Xiaomi/gimi-illustration-skill gimi-illustration --agent gemini-cli

Project scope by default (.agents/skills/); add --scope user for a personal install.

Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/

Paste into Gemini CLI

Install the "gimi-illustration" agent skill from https://github.com/GiMi-Xiaomi/gimi-illustration-skill/tree/main into .gemini/skills/gimi-illustration/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "gimi-illustration", then confirm the skill loads.

Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

GitHub CLI

$ gh skill install GiMi-Xiaomi/gimi-illustration-skill gimi-illustration

Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).

skills CLI

$ npx skills add GiMi-Xiaomi/gimi-illustration-skill --skill gimi-illustration -a github-copilot

Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.

GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/

Paste into GitHub Copilot

Install the "gimi-illustration" agent skill from https://github.com/GiMi-Xiaomi/gimi-illustration-skill/tree/main into .github/skills/gimi-illustration/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "gimi-illustration", then confirm the skill loads.

GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add GiMi-Xiaomi/gimi-illustration-skill --skill gimi-illustration -a opencode

OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.

GitHub CLI

$ gh skill install GiMi-Xiaomi/gimi-illustration-skill gimi-illustration --agent opencode

Project scope by default (.agents/skills/); add --scope user for a personal install.

OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/

Paste into OpenCode

Install the "gimi-illustration" agent skill from https://github.com/GiMi-Xiaomi/gimi-illustration-skill/tree/main into .opencode/skills/gimi-illustration/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "gimi-illustration", then confirm the skill loads.

OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

Facts

Skill name: gimi-illustration
GitHub stars: 703
Token cost: ~1.7k tokens
SKILL.md length: 552 words
Files: 55 (incl. references, assets)
Skills in repo: 1
Repo updated: Aug 3, 2026
First seen: Oct 7, 2026
Licence: MIT

At a glance

Works in 7 steps: 自定义 IP（按需） → 意图确认 → 5 · 当前参考解析（IP 与无 IP） → …
Is enrolling a custom IP: do NOT generate a poster
SKILL.md covers ⚠ 路由分流（先读 · 防 ImageGen 抢跑）, 触发词, 参考文件（按需加载） and 生图能力, plus 3 more sections
Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Gimi Illustration is an agent skill from GiMi-Xiaomi/gimi-illustration-skill. Multi-style illustrations from Chinese article text (配图 / illustration): quirky sketch, warm storybook, product proposal; OR custom IP enrollment (自定义 IP / 录入 IP / 上传形象 / 新建 IP). When the user is enrolling a custom IP: do NOT generate a poster or illustration first — load references/ip/template.md and run Step 0 (plan confirm before freeform image gen). Use for 配图, 给文章配图, 帮我配图, illustration, 自定义 IP, 录入 IP, 上传形象, 新建 IP, 怪诞手绘, 暖调绘本, 产品方案.

Its SKILL.md is about 1.7k tokens, which your agent loads only when the skill is triggered. The skill folder holds 60 other files, including reference files and assets (for example `CHANGELOG.md`, `IP-NOTICE.md` and `README.md`).

It sits in Frontend & Design, covering Image generation. It works with Storybook. The repository describes itself as: 文章/口播配图 skill ，支持自定义 IP、调尺寸、配置标题、换风格. The licence is MIT.

When your agent uses it

Is enrolling a custom IP: do NOT generate a poster
Illustration first — load references/ip/template.md and run Step 0 (plan confirm before freeform image gen)

Example prompts

“/gimi-illustration”

Workflow steps

7 steps, taken from the step headings in SKILL.md.

自定义 IP（按需）
意图确认
5 · 当前参考解析（IP 与无 IP）
消化内容 + 出策略
组装 Prompt + 生图
QA 质检
交付

What it can do on your machine

Read from SKILL.md and the folder at commit 0f602ea. It shows what the files ask for, not the result of running them.

Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
No scripts in the folder and no shell commands in SKILL.md.
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Gimi Illustration loads about 1.7k tokens when it runs, and up to ~28k if it reads all its reference files. Until then it costs about 115 tokens; SKILL.md has 552 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it: ~115
When it runs · the whole SKILL.md, loaded when a task matches: ~1.7k
With references · SKILL.md plus every file in references/, read only if the agent opens them: ~28k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from GiMi-Xiaomi/gimi-illustration-skill at commit 0f602ea, republished under its MIT licence (© GiMi-Xiaomi). 552 words, ~1,688 tokens.

Download SKILL.mdSave it as .claude/skills/gimi-illustration/SKILL.md (or your agent's skills folder). This skill also uses 54 other files; get the full folder from GitHub.

name: gimi-illustration
description: Multi-style illustrations from Chinese article text (配图 / illustration): quirky sketch, warm storybook, product proposal; OR custom IP enrollment (自定义 IP / 录入 IP / 上传形象 / 新建 IP). When the user is enrolling a custom IP: do NOT generate a poster or illustration first — load references/ip/_template.md and run Step 0 (plan confirm before freeform image gen). Use for 配图, 给文章配图, 帮我配图, illustration, 自定义 IP, 录入 IP, 上传形象, 新建 IP, 怪诞手绘, 暖调绘本, 产品方案.

gimi-illustration · 配图 Skill

Public 3.0 = 产品 v3.0（多风格：怪诞手绘 / 暖调绘本 / 产品方案 + 自定义 IP）

换窗口：有 docs/ 时先读 docs/INDEX.md → docs/roadmap.md；执行以本文件 + references/ip/_template.md 为准

⚠ 路由分流（先读 · 防 ImageGen 抢跑）

触发含 自定义 IP / 录入 IP / 上传形象 / 新建 IP（可附图）时：

立刻加载 references/ip/_template.md，走 Step 0；不要先当「配图 / 海报」生图
本回合禁止调用任何生图工具（image_gen / GenerateImage / ImageGen 等），直到用户已回答「按这个形象方案录入吗？」
- 半身/全身相同：首轮只出形象方案 + 收尾问；零例外
- 确认后：半身才允许 0.2b 全身设定稿；再 0.4 校准
禁止把录入理解成「生成一张 IP 主题海报 / 原创插画海报」
细则与话术以 _template 为准；本段与之冲突时，以本段路由为准（先拦抢跑）

触发仅为 配图 / illustration 等、且非录入 IP → 走下方 Quick Mode / Step 1–5。

输入文章内容，生成配图。默认风格「怪诞手绘」（quirky-sketch），可切换已录入风格（如「暖调绘本」）；IP 为 gimi / none / 自定义实例，默认 16:9。

触发词

配图：配图 / 给文章配图 / 帮我配图 / illustration
自定义 IP：自定义 IP / 录入 IP / 上传形象 / 新建 IP

参考文件（按需加载）

文件	何时加载
`docs/INDEX.md`	换窗口文档导航（先读）
`docs/roadmap.md`	换窗口工作台 / 要改规则时
`docs/spec.md`	产品规划（防跑偏）；非执行协议
`references/styles/{$STYLE}.md`	生图 / 校准 / 风格 QA 时
`references/styles/_template.md`	批量录入新风格时（非配图热路径）
`references/composition-patterns.md`	出策略时（非 Step 0）
`references/visual-promise.md`	Step 2 判视觉承诺 / 结构·外参·域物件（SSOT；细则不抄进本文件）
`references/shot-config.md`	出策略时（非 Step 0）
`references/prompt-template.md`	组装 prompt 时
`references/qa-checklist.md`	质检时
`references/ip/_template.md`	自定义 IP / Step 0（优先于生图）
`references/ip/gimi/ip.md`	`$IP` = gimi
`references/ip/none.md`	`$IP` = none
`references/ip/{id}/ip.md`	`$IP` 为已存在的自定义 id

生图能力

调用当前 AI 软件内置的生图工具（如 Codex @Image Gen、Cursor GenerateImage 等）。Skill 负责策略、prompt 组装、参考图附件、QA；生图由平台工具执行。接口语义：文本 prompt + 可选图片附件。

Step 0 录入中： 遵守文首「路由分流」；未满足允许条件前不得调用生图工具。

平台适配优先级：

支持显式图片附件的平台：把参考图作为 image reference 附件传入。
Codex image_gen：已有 IP 的正式配图一律传「设定图 + 当前 (IP, STYLE) 校准图」；先跑 Step 1.5，缺任一张不得生正式图；再 image_gen
不支持图片输入且无法把图片放入上下文时，暂停说明原因；已有 IP 不得降级为仅文字锚点的正式配图。

核心变量

变量	默认值	说明
`$STYLE`	`quirky-sketch`	默认怪诞手绘；可按主人语言切换为已录入风格（见 Step 1）
`$RATIO`	`16:9`	宽高比；用户可指定任意比例
`$IP`	`gimi`	`gimi` / `none` / 其他：若存在 `references/ip/{id}/`（非 `_template`）则加载该实例
`$REF_MODE`	解析得出	`dual` / `style-single` / `none`；已有 IP 缺当前风格校准时为 `lazy` 门禁，不是可交付参考模式
`$TITLE`	无	用户要求时顶部手写标题
`$COUNT`	自动推断	配图张数
`$OUTPUT_SIZE`	无	可选；仅用户明确要求具体像素时记录

推荐像素（非强制）： 16:9 → 建议 1920×1080；3:4 → 建议 1080×1440。

工作流

Quick Mode

用户已给正文且变量可推断 → 不多问，直接生图。以下情况才确认：用户说「先别生图」、变量冲突、新 IP 尚无 ip.md（先走 Step 0）、或 Step 1.5 判为 lazy。 Step 0 进行中：关闭 Quick Mode（不得跳过确认话术）。
录入触发词出现时：不适用 Quick Mode 生图。

Step 0 · 自定义 IP（按需）

用户要新建/录入 IP 时：加载 references/ip/_template.md，严格按其中 Step 0 填槽（用户只需附图 +「录入 IP」，不必写 Step 编号）。
双层： 协议走 0.1–0.5；对用户只用「用户层话术」（见 _template），称当前 $STYLE 的 display_name（录入默认多为「怪诞手绘」），不暴露 Step 编号。

硬门禁（0.5 完成前）——与文首路由分流一致：

入口（半身/全身相同）：首轮只出形象方案 +「按这个形象方案录入吗？」；一律禁生图；禁海报式成稿
禁止配图抢跑：不得进 Step 1–5；不得写 shot-config.md
出图后必须提问：同轮追加收尾问；图 ≠ 闭合（仅 0.2 确认之后才允许生图）
半身：确认后 0.2b 须头到脚；落盘 reference-character.png 一律 full
0.4 标准风格样板镜（读 styles/{$STYLE}.md）；「像吗？」未闭合前不得进配图

examples/{$STYLE}/ 空且要用该风格配图时 → lazy 补 0.4（已启用，见 _template 多风格细则）。
完成后才进入配图；配图时 $IP={id}。

Step 1 · 意图确认

记录：$RATIO · $IP · $STYLE · $TITLE · $COUNT ·（可选）$OUTPUT_SIZE

Quick Mode 推断：竖版/小红书 → 3:4；否则 → 16:9。

$STYLE 解析（主人语言，不要求用户写 id）：

扫描 references/styles/*.md（排除 _template.md）的 display_name / aliases
明确命中 → 锁定对应 id
软命中 → 一句确认（例：「你是说「暖调绘本」那种纸纹暖色吗？」）
未命中或「换个风格」→ 人情味菜单：每项 display_name + tagline（默认可标「常用」）
未提风格 → quirky-sketch；本会话首次配图追加一句轻提示（不打断）：
「这次先用怪诞手绘。若想换成暖调绘本（纸纹暖色、旅行手帐感），直接说就行。」
用户描述未录入风格 → 不现场发明；说明目前仅已录入风格，问用默认或以后再录

口语优先；$STYLE=… 仅进阶。对用户称 display_name，不说路径。

Show full SKILL.md (205 more words)Show less

Step 1.5 · 当前参考解析（IP 与无 IP）

在进入 Step 2、写 shot-config.md 或生成正式图之前，对当前 (IP, STYLE) 只判一次：

$IP=none 且当前风格「无 IP 校准策略」为 required style-single → 必须按 references/ip/none.md 选择 assets/none/examples/{$STYLE}/ 中 1 张，记为 $REF_MODE=style-single；缺目录 / 缺选图是发布包资产问题，停止并修复，不向用户索图、不降级为纯文本。
$IP=none 且当前风格没有 required 策略 → $REF_MODE=none，保持既有 none 行为；若该风格声明 optional 且有资产，可按 none.md 选 1 张作质量收敛。
已有全身设定图且 assets/ip/{$IP}/examples/{$STYLE}/ 至少有 1 张可用校准图 → $REF_MODE=dual，可进 Step 2。
已有全身设定图但当前风格校准目录为空 / 缺文件 → $REF_MODE=lazy；只按当前风格 0.4 生成样板，QA 后问「像吗？」；禁止写 shot-config.md、生成正式图、或附其他风格校准图充当双参考。用户催图也不降级；应说明「这个风格的样板还没定，确认样板后才能正式配图」。
设定图缺失或不可用 → 路由到现有 IP 录入；不得 single 降级或生成正式图。

用户确认「像」后才将该样板落盘到当前 $STYLE 目录；后续正式请求重新解析为 dual。已有 IP 的正式配图不允许 single，必须 dual。 IP 文件不维护全局 ref_mode；当前风格的目录事实是唯一依据。

Step 2 · 消化内容 + 出策略

仅 $REF_MODE 已被 Step 1.5 判为 dual / style-single / none 时可进入。策略：composition-patterns.md · 视觉承诺：必载 visual-promise.md（SSOT）· 落盘：shot-config.md。

加载 composition + shot-config + visual-promise
按 composition Step 2 决策流程（二轴 → 创意生成法八步 → 构图库）
每张 shot 填视觉承诺全家桶（visual-promise.md）+ density / 阅读顺序（composition「阅读动线语法」；scenic 强制）
外参参考：按 visual-promise §B（主锚预算 + 默认特征锁不附图）；失败/超时静默降级 意象（不问用户）
Quick Mode：默认落盘进 Step 3；用户说「先看策略」时先展示摘要

Step 3 · 组装 Prompt + 生图

加载 styles/{$STYLE}.md、prompt-template.md、对应 ip/*.md
只消费 Step 1.5 已解析的 $REF_MODE：
- dual → 设定图 + examples/{$STYLE}/ 中 1 张校准图（两张都存在才可组装）
- style-single → $IP=none，必传 assets/none/examples/{$STYLE}/ 中按 none.md 选定的 1 张；prompt 明示只匹配画法，不复制构图、标题、物件排列或业务内容
- none → 仍按 none 规则（无角色设定图，也无 required 风格资产）
- 参考图未进上下文，不得声称已执行有 IP 流程
按 visual-promise.md / prompt-template：追加承诺锁；外参参考 默认不附地标图（特征进 SUBJECT）；仅「外参附图=是」才附；scenic 追加阅读动线意涵（composition SSOT）
填充变量槽，调用生图工具（IP_DESC 在 STYLE_DNA 之前，见 prompt-template.md）

Step 4 · QA 质检

加载 qa-checklist.md（承诺 → visual-promise；动线 → composition「阅读动线语法」）。$OUTPUT_SIZE 有值时 → 等比缩放 + 白底留白，禁止裁切。不通过 → 重试，最多 2 次。

Step 5 · 交付

outputs/{YYYYMMDD}-{slug}/
  shot-config.md      ← 配图策略 + 插入位置（与图片同时交付）
  01-{slug}.png
  02-{slug}.png       ← 多张时继续编号

版本

Public / 产品 3.0 · 三风格可选 + 自定义 IP · visual-promise SSOT

Files

SKILL.md and 54 other files (references, assets) in the repository root of GiMi-Xiaomi/gimi-illustration-skill.

SKILL.md
.gitignore
CHANGELOG.md
IP-NOTICE.md
LICENSE
README.md
assets/ip/gimi/.gitkeep
assets/ip/gimi/examples/product-proposal/gimi-ai-change-chaos.png
assets/ip/gimi/examples/product-proposal/gimi-desk-shoulder-fatigue.png
assets/ip/gimi/examples/product-proposal/gimi-exercise-modal-focus.png
assets/ip/gimi/examples/product-proposal/gimi-writing-rules-checklist.png
assets/ip/gimi/examples/quirky-sketch/gimi-ai-change-chaos.png
assets/ip/gimi/examples/quirky-sketch/gimi-desk-shoulder-fatigue.png
assets/ip/gimi/examples/quirky-sketch/gimi-exercise-modal-focus.png
assets/ip/gimi/examples/quirky-sketch/gimi-writing-rules-checklist.png
… and 40 more

Open the folder on GitHubat commit 0f602ea

Compare with similar skills

Gimi Illustration next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Gimi Illustration compared with similar skills
Skill	Stars	Used in	Tokens	Auto-check	Licence	Repo updated
Gimi Illustration this skillGiMi-Xiaomi/gimi-illustration-skill	703	—	~1.7k	Automated safety check: Pass	MIT	2 mo ago
Story HTML Publisherhassancs91/claude-image-generation	102	—	~1.7k	Automated safety check: Pass	MIT	1 mo ago
Scene Splitterhassancs91/claude-image-generation	102	—	~2.3k	Automated safety check: Pass	MIT	1 mo ago
Ky Design To HTMLKyrieCheungYep/ky-design-to-html-skill	163	—	~2.8k	Automated safety check: Pass	None	4 mo ago
Qovery UIQovery/console	227	—	~2k	Automated safety check: Pass	GPL-3.0	yesterday
Frontend Design Proclaudekit/frontend-design-pro-demo	278	—	~1.4k	Automated safety check: Pass	None	10 mo ago

Similar skills

Story HTML Publisher
hassancs91/claude-image-generation
Final step of the AI Storybook pipeline. An agent skill from hassancs91/claude-image-generation.
102 GitHub stars~1.7k tokensUpdated 1 mo ago
Frontend & DesignAuto-check passed
Scene Splitter
hassancs91/claude-image-generation
Splits a plain English story into a numbered list of SCENES — each scene being one moment that gets exactly one illustration AND one narration clip downstream.
102 GitHub stars~2.3k tokensUpdated 1 mo ago
Frontend & DesignAuto-check passed
Ky Design To HTML
KyrieCheungYep/ky-design-to-html-skill
Convert visible UI references such as screenshots, exported mockups, Figma frame images, SaaS empty states, dashboards, and landing page screenshots into high-quality HTML/CSS by decomposing layout…
163 GitHub stars~2.8k tokensUpdated 4 mo ago
Frontend & DesignAuto-check passed
Qovery UI
Qovery/console
Design review and implementation guidance for the Qovery Console.
227 GitHub stars~2k tokensUpdated yesterday
Frontend & DesignAuto-check passed
Frontend Design Pro
claudekit/frontend-design-pro-demo
Creates jaw-dropping, production-ready frontend interfaces AND delivers perfectly matched real photos (Unsplash/Pexels direct links) OR flawless custom image-generation prompts for hero images…
278 GitHub stars~1.4k tokensUpdated 10 mo ago
Frontend & DesignAuto-check passed
Omd Media
kwakseongjae/oh-my-design
Brand-consistent asset sets from DESIGN.md — hero, section illustrations, icon sets, OG image — generated through whichever channel the user actually has (grok build imagegen, Codex $imagegen…
533 GitHub stars~1.1k tokensUpdated 10 days ago
Frontend & DesignAuto-check passed

Works with

Storybook

Questions about Gimi Illustration

What does Gimi Illustration do?

Multi-style illustrations from Chinese article text (配图 / illustration): quirky sketch, warm storybook, product proposal; OR custom IP enrollment (自定义 IP / 录入 IP / 上传形象 / 新建 IP). Gimi Illustration is an agent skill from GiMi-Xiaomi/gimi-illustration-skill. Multi-style illustrations from Chinese article text (配图 / illustration): quirky sketch, warm storybook, product proposal; OR custom IP enrollment (自定义 IP / 录入 IP / 上传形象 / 新建 IP).

When should I use Gimi Illustration?

Gimi Illustration fits situations like: is enrolling a custom IP: do NOT generate a poster; illustration first — load references/ip/template.md and run Step 0 (plan confirm before freeform image gen).

How do I install Gimi Illustration in Claude Code?

Run `npx skills add GiMi-Xiaomi/gimi-illustration-skill --skill gimi-illustration -a claude-code`. Or copy the skill folder (the GiMi-Xiaomi/gimi-illustration-skill repository) into .claude/skills/gimi-illustration in your project. Claude Code loads it when a task matches its description.

How do I install Gimi Illustration in Codex?

Run `npx skills add GiMi-Xiaomi/gimi-illustration-skill --skill gimi-illustration -a codex`. Or copy the skill folder (the GiMi-Xiaomi/gimi-illustration-skill repository) into .agents/skills/gimi-illustration in your project. Codex loads it when a task matches its description.

Can I use Gimi Illustration in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add GiMi-Xiaomi/gimi-illustration-skill --skill gimi-illustration -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/gimi-illustration, .gemini/skills/gimi-illustration, .github/skills/gimi-illustration and .opencode/skills/gimi-illustration in your project.

What does Gimi Illustration need to run?

SKILL.md names no scripts, command-line tools or credentials: Gimi Illustration is instructions for the agent only.

Does Gimi Illustration access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Gimi Illustration safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Gimi Illustration use?

Gimi Illustration is published under the MIT licence (from the LICENSE file in the skill folder). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Gimi Illustration use?

About 1.7k tokens (SKILL.md is roughly 6.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 26k tokens, read only when the agent opens those files.

What are the alternatives to Gimi Illustration?

Skills that share tags, products or a category with Gimi Illustration: Story HTML Publisher (hassancs91/claude-image-generation, 102 stars), Scene Splitter (hassancs91/claude-image-generation, 102 stars), Ky Design To HTML (KyrieCheungYep/ky-design-to-html-skill, 163 stars) and Qovery UI (Qovery/console, 227 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Gimi Illustration?

GiMi-Xiaomi (a GitHub user) maintains it in GiMi-Xiaomi/gimi-illustration-skill, which has 703 GitHub stars. The repository was last updated on August 3, 2026.

Source: GiMi-Xiaomi/gimi-illustration-skill on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.

Install Gimi Illustration

Facts

At a glance

What it does

When your agent uses it

Example prompts

Workflow steps

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Context cost

Safety

SKILL.md

gimi-illustration · 配图 Skill

⚠ 路由分流（先读 · 防 ImageGen 抢跑）

触发词

参考文件（按需加载）

生图能力

核心变量

工作流

Quick Mode

Step 0 · 自定义 IP（按需）

Step 1 · 意图确认

Step 1.5 · 当前参考解析（IP 与无 IP）

Step 2 · 消化内容 + 出策略

Step 3 · 组装 Prompt + 生图

Step 4 · QA 质检

Step 5 · 交付

版本

Files

Compare with similar skills

Similar skills

Story HTML Publisher

Scene Splitter

Ky Design To HTML

Qovery UI

Frontend Design Pro

Omd Media

Related

Questions about Gimi Illustration