ip-diagram-creator skill
A skill that turns your photo or description into a hand-drawn character and uses it for diagrams, knowledge cards and slides
- 263 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- The skill handles your photos, profile screenshots and descriptions, which go to the model and the image generator
- Rights must be checked on the result: the skill does not replace a check of trademarks, image rights and advertising rules
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/ip-diagram-creatorDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The skill handles your photos, profile screenshots and descriptions, which go to the model and the image generator
- Rights must be checked on the result: the skill does not replace a check of trademarks, image rights and advertising rules
This is third-party code. Review the repository files before installing.
What it does
ip-diagram-creator builds a signature character in a minimalist drawn style for an author or expert and then illustrates texts, courses and talks with it. You provide the source material yourself: authorised photos, a profile page screenshot, a short bio or an existing character sheet. The principle: first the character is fixed, then the content is broken down, and only then are diagrams drawn. The first stage yields three fixed assets: the main portrait, a sheet of character rules and a sheet of motions, expressions and small-scale scenes. The skill then reads an article, script or single topic and suggests how many images are needed and of what type. Modes are listed in a table: agent recommendation, a knowledge card for a method or comparison, a drawn illustration for one idea, a shot list for a long text and a slide mode for a talk. In slide mode the skill first prepares a director's plan card: the type of each page, information density and how often the character appears. It then makes one or two trial slides and generates the whole series in batches, checking it against a contact sheet. Knowledge cards by default include 2 to 6 executor agents who move cards, hand in results and flag problems while the character explains. The skill installs with `npx skills add`. If the agent has no image generation tool it outputs ready prompts for GPT Image 2 or `image_gen` plus repair prompts and does not pretend the picture exists.
Who it is for. For authors, teachers, consultants and content teams who need a recognisable drawn character for illustrations, knowledge cards and slides.
Good fit when
- You need a series of illustrations for an article with the same character and want to pick the count and type before generating
- A method or process must become a knowledge card with steps, comparison and flagged problems
- A talk or course is being prepared: slide mode makes a page plan, trial slides and a final contact sheet for review
- You have only a photo and a profile page and need a stable character for long-term use
- You have no image generator but need precise prompts for GPT Image 2 with spare repair prompts
Not a fit when
- You need documents or portrait photography: the skill does not reproduce looks exactly and does not make ID or commercial portraits
- The source materials are someone else's: photos without permission, screenshots of other people's pages and private messages, client data
- You need a check of trademarks, image rights and advertising rules: the skill does not replace it
- You want arbitrary illustration styles: the skill is built around one minimalist drawn style
Example request
Here is my photo and my expert page description. Propose a character, then read my article and say how many illustrations it needs, but do not generate anything yet.Limitations
- The skill works best where the agent can read images, thinks at a high level and has an image generation tool. Without one you get prompts to paste into a generator by hand.
- It does not guarantee a likeness to a real face and is not suited to ID photos or commercial portraits.
- The skill does not use photos without permission, screenshots of other people's pages and private messages, client materials or references that cannot be distributed.
- Decisions on trademarks, image rights, platform rules and advertising stay with a person.
- Personal character assets you have approved do not go into the skill's public package.
How to disable. Delete the `ip-diagram-creator` folder from the agent's skills directory (`~/.claude/skills/` or the project's `.claude/skills/`). Images and character sheets you created stay with you and are removed separately.
FAQ
How do I install ip-diagram-creator?
Run `npx skills add https://github.com/haloshin/ip-diagram-creator` for the current project or add `-g` for a global install. You can also download the repository and put the folder in the agent's skills directory, which will read `SKILL.md`.
What does the skill create first?
Three character assets: the main portrait, a rules sheet and a sheet of motions, expressions and small-scale scenes. Only after you confirm them does the skill move to illustrations.
How do I make slides for a talk?
Provide a topic, outline, script or an old deck. The skill drafts a director's plan card, makes one or two trial slides and, once you approve, generates the whole series in batches.
What if the agent has no image generation?
The skill outputs full prompts and spare repair prompts you can paste into GPT Image 2 or another generator. It does not claim a picture exists when it has no tool.
Which materials must I not use?
Photos without permission, other people's screenshots, private messages, client data and references that cannot be distributed. The skill is built for materials you authorised yourself.
Discussion
No questions or reviews yet.
Related
SkillsA skill that generates extremely simple, cute mascot characters for logos: rounded forms, three colors, lower-corner composition
A skill that draws original print-style illustrations: one idea per frame, played out by a recurring mascot
A Codex skill that redraws your photo as a watercolor illustration on white paper, with wide margins and a handwritten caption
A skill that takes an agent from a generated picture to a game asset: sprites, video-to-frames animation and pre-release checks
A skill that draws interactive isometric illustrations in SVG, with WebGL shaders and React output, explaining an idea as a working machine
A loop skill: the agent generates a target screenshot, builds toward it, and uses a critic subagent to close the gap until they match
Comparison with similar tools
Entries that share the main search query, and how they differ.
| Entry | Official | Repository stars | Main difference |
|---|---|---|---|
| ip-diagram-creator (this entry) | no | 263 | |
| IP as Logo | no | 6,280 | creates very simple mascot characters for logos, while here the character leads diagrams, knowledge cards and slides |
| gpt-image2-ppt-skills | no | 1,404 | draws decks with gpt-image-2, while here the focus is a steady author character across all materials |
README in short
The page in Chinese gives install, a first run and a table of modes, then shows what you get: three character assets, diagram types and a slide mode with examples. It ends with limits plus input and output data.
README badge
Original SKILL.md
--- name: ip-diagram-creator description: 个人 IP 图解创作器。用你的 IP 形象，生成专属内容图解。用于把用户授权提供的照片、主页截图、简介、账号资料或角色档案转成稳定的极简手绘图解角色，并继续生成长文配图、图文号知识卡、手机海报、方法拆解图、分镜故事图、IP 演讲 PPT、直播分享课件和全图化手绘 PPT 页面等内容视觉。它强调 IP 主角 + 执行 Agent 的知识图解系统：用户 IP 做主讲、拆解、批注和调度，执行 Agent 做搬卡片、反馈、试错、标记风险和递交结果。用户想做个人 IP 形象、用自己的形象做知识图、图文号知识卡、文章配图、课程图解、内容海报、方法拆解图、IP 演讲 PPT、直播分享课件，或上传照片/主页截图/简介希望做内容视觉时，使用本 Skill。 license: MIT --- # 个人 IP 图解创作器 用你的 IP 形象，生成专属内容图解。 底层原则：先定角色，再拆内容，最后生成图解。 本 Skill 不是头像生成器，也不是自由风格生图器。它的目标是把用户授权提供的个人材料转成一套可长期复用的极简手绘图解角色，再用这个角色和执行 Agent 一起参与内容表达。 ## 推荐运行环境 推荐在支持图片读取、较高推理强度和图像生成工具的 Agent 环境中使用。这样可以直接完成图片读取、角色建档、三张角色资产生成、内容图解生成和返修。 如果当前环境支持 `image_gen`、GPT Image 2 或其他图像生成工具，可以在用户明确说“生成 / 出图 / 直接做图”时直出图片。建议在角色建档、长文拆解、知识卡设计和角色一致性返修时使用较高推理强度。 如果当前环境没有图像生成工具，不要假装已经出图。改为输出可复制到图像生成工具的完整 prompt，并附返修 prompt。 ## 环境判断 每次开始前先判断当前能力： 1. 如果有图像生成工具： - 用户明确说“生成 / 出图 / 直接做图”时，可以直接生成。 - 每张图单独生成，不要把多张图拼成一张，除非用户明确要卡组总览。 - 调用前仍要构造完整 prompt，便于返修和复用。 2. 如果没有图像生成工具： - 不直接声称生成了图片。 - 输出可复制到常见图像生成模型的完整 prompt。 - 同时输出返修 prompt。 - 如果用户需要多张资产，分别输出 `01_角色主锚图`、`02_角色规范说明图`、`03_动作_表情_小比例场景扩展图` 三个 prompt。 3. 如果当前模型或环境不能读取图片： - 先让用户补充外形描述，或要求上传可读取图片。 - 不能凭空捏造强相似角色。 4. 如果用户使用低推理或非图像模型： - 优先交付结构化角色提取卡、内容确认卡和完整 prompt。 - 不强行承诺复杂图文编排、角色一致性或图片已生成。 ## 用户侧语言 对用户使用这些说法：