Yingzao skill
A skill that turns photos of buildings, streets and local culture into art-directed editorial posters via an image model
- 532 ★ on the repository
- Last commit
Extended access
Why this level
- Runs Python scripts and reads local images
- Calls an external image generation model
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/yingzaoDetects 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.
- Runs Python scripts and reads local images
- Calls an external image generation model
This is third-party code. Review the repository files before installing.
What it does
The skill takes real photos of architecture, interiors, objects and food and composes an editorial poster rather than applying a filter. It first reads the composition and identifies the shot, a facade, a low angle, a diagonal or a detail, fixes one genuinely distracting framing issue and protects the building's identity features. Then it picks a compatible reference image and designs Chinese display type and the spatial relation of text to image. On the final step it hands the source, the reference and a sparse layout underlay to an image model so it repaints the whole scene. There are checks before the model call and one diagnostic after.
Who it is for. For designers and visual content authors who need posters from real photos of places and culture.
Good fit when
- You need an editorial poster from a photo of a historic building or street
- You need to merge several photos into one scene
- You need Chinese display type tied to the main subject on a poster
Not a fit when
- You need a plain ad or product image for a shop
- You only need exposure fixes, retouching or a sky swap
- There is no real place or subject to base it on and you want an invented building
Example request
Use $yingzao to make a 3 by 4 editorial poster from this old-town photoLimitations
The skill only works with an agent that can read local images and call an image model; the README mentions GPT Image, and access to such models may be blocked in some regions without extra tools. It needs Python 3.10 and Pillow, NumPy, OpenCV and fontTools, with a dependency check before the first run. The interface and references are mostly Chinese.
How to disable. Delete the yingzao folder from your agent's skills directory.
FAQ
Does the skill draw the poster itself?
No, an image model builds the final picture; the skill prepares the source, reference, layout and checks.
What besides the skill do I need?
Python 3.10 and Pillow, NumPy, OpenCV and fontTools; a dependency-check script ships with it.
Discussion
No questions or reviews yet.
Related
SkillsSkill that turns a photo into an original 3:4 art poster: scene analysis, a fitting medium, palette and typography
A Codex skill that turns a photograph into a vertical composition with an abstract memory panel and an English title
A skill that makes an editorial poster from a photo: the shot stays unchanged, with an abstract panel and a local title beside it.
A skill that generates one-ink and two-ink editorial posters: paper substrate, halftone, negative space and restrained type
A Codex skill that turns a theme, phrase or photo into a paper-zine editorial poster, a ready image prompt or a reusable set of style rules
Portrait skill that makes a snapshot cinematic while keeping face and scene, or builds an original series from scratch.
README in short
The README describes a skill that turns photos of architecture and local culture into editorial posters through an image model. Installation is a single npx skills add command, and there is a text you can pass to an agent to install it itself. The workflow is laid out: photo and fact preflight, four creative domains, one dominant reference, type and sparse-layout design, whole-image generation and one diagnostic afterward. Running it needs Python 3.10 and Pillow, NumPy, OpenCV and fontTools; outputs are written to the working directory, not into the skill.
README badge
Original SKILL.md
--- name: yingzao description: Transform real Chinese architecture and place-based cultural photos into art-directed editorial posters, integrated multi-photo scenes, and optional source comparisons. Use for historic buildings, vernacular streets, culturally rooted shops, interiors, food, travel covers, and place montages; do not use for generic advertising, routine retouching, or invented places. --- # Yingzao · 营造 把真实建筑与在地文化照片转译为有主体处理、主动背景、字形设计和空间互动的编辑海报。先从照片提出创意命题，再让一张兼容的主导参考强化它；Token 负责检索与机制追踪，不能替代设计判断。 ## 核心合同 - 高风格化任务固定使用：Image 1 校正原图、Image 2 一张主导参考实图、Image 3 稀疏中性排版垫图。生成前必须由 `scripts/prepare_generation.py` 编译成一次调用清单，不得互相替代、改序或在调用时重新概括提示词。 - 所选 Recipe 必须保存为 `analysis/recipe.json`，再把全部启用 Token 编入 `analysis/design-plan.json`。 - 创意命题必须先于参考检索。 ## 输入确认 从用户消息提取：地点/对象与必要文案、海报数量、是否要原图对照拼图。 ## 工作流 ### 0. 建立安全输入 1. 运行 `python3 scripts/check_dependencies.py`。 2. 产物写入调用者工作目录的 `output/yingzao/<run-id>/`，不得写入 Skill 包。 3. 每张来源运行 `scripts/photo_preflight.py`，判定 `hero / support / reject`。 4. 事实按 `VERIFIED / OBSERVED / USER-CONFIRMED / UNCONFIRMED` 分级；只有前三类进入海报。