HandwritingGenerator skill
A local handwriting image generator with CLI and MCP: an agent picks font and paper, saves a preset and renders text offline.
- 138 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Handwritten images are easy to pass off as hand-written work, which is the user's responsibility.
- The agent runs local commands and writes files into `outputs`.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/handwritinggeneratorDetects 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.
- Handwritten images are easy to pass off as hand-written work, which is the user's responsibility.
- The agent runs local commands and writes files into `outputs`.
This is third-party code. Review the repository files before installing.
What it does
HandwritingGenerator, also called Xingjian, turns text into an image with a handwritten look. An agent selects paper and font, tunes margins, size, letter spacing and ink colour and saves a PNG, JPEG or WebP. The layout can be stored as a JSON preset and reused with other text. The repository has two implementations: Rust with an egui interface, nine bundled fonts and GUI, CLI and MCP modes, and Python with PyQt6. The `handwriting-generator` skill explains `cli.py assets` and `cli.py render` to an agent. With MCP connected, `list_assets` and `render_handwriting` are available. No API key is needed and text is processed locally.
Who it is for. People who need cards, letters and notes as a handwritten-style image. Works with Codex or Claude Code with terminal access.
Good fit when
- A card or letter as a handwritten page.
- One layout is reused for several texts through a preset.
- An agent must choose a font and paper from what is really available.
- The image is made without sending text to external services.
Not a fit when
- A handwritten piece would be passed off as hand-written where that is forbidden.
- A long text is needed without splitting, because pages are not cut automatically.
- Forging a particular person's handwriting: the project has only bundled fonts.
Example request
List available fonts and paper, then render this greeting as handwriting and save a preset for later ones.Limitations
Content outside the drawing area is cropped, there is no automatic pagination, so long texts must be split or set smaller. Existing output files are not overwritten by default. A preset contains the text itself, so replace it on reuse. A handwritten look does not prove a person wrote by hand. The Python version needs Python 3.10 or newer and the Rust version is built from source. Interface and description are Chinese.
How to disable. Delete `.agents/skills/handwriting-generator` and the MCP entry. Clean the `outputs` folder of finished images by hand.
FAQ
Is an API key needed?
No. Text and images are processed locally, as the README states directly.
Which export formats are supported?
PNG, JPEG and WebP. PNG keeps a transparent background and the size matches the source paper.
How do I reuse one style for different texts?
Save a preset with `--save-preset`, apply it with `--preset` and supply new text with `--text` or `--text-file`.
What about long text?
Enlarge the area, reduce the font size or split the text across several images. There is no automatic pagination.
Which versions are required?
The Python version needs Python 3.10 or newer, ideally 3.12. The Rust version is built from source with `cargo run`.
Discussion
No questions or reviews yet.
Related
SkillsSkill, MCP server and CLI that let any agent call Codex image generation, after picking a style from several directions
A set of skills for agents to generate and edit images, video and audio through the muapi.ai service and muapi-cli
Image generation and editing from any agent: the imagegen-agent CLI and a skill with the Codex prompt library for compatible APIs
MCP server for image generation and editing: Gemini, OpenAI and Seedream with automatic prompt enhancement
A skill that adds image generation to Claude Code and Cursor through the Codex CLI on a ChatGPT subscription, no API key.
A set of skills for Claude Code and Codex that assemble detailed image prompts from an element library and composition schemes
README in short
One renderer serves the window, CLI and MCP, so agent and human output match. Fonts and backgrounds ship in the repository.