An automated diagramming skill for AI coding agents to design, configure, and output production-ready hand-drawn Excalidraw charts.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install hermes-excalidraw
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install hermes-excalidraw using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The Excalidraw Diagram Skill enables AI agents to generate structured, professional-grade hand-drawn charts without relying on bloated external rendering tools or cloud API accounts. It is designed to work natively as part of the broader Openclaw Skills ecosystem, giving developer agents an elegant way to express complex software architectures, workflow pipelines, and data topologies in clear, human-readable diagrams.
By programmatically generating standard Excalidraw JSON files, this skill maintains extreme layout precision. It enforces structural grid arrangements, uniform element sizes, and strict text-to-box sizing calculations to eliminate layout bugs like overlapping text, clipped labels, or crossing connector lines. The generated files can be imported directly into Excalidraw or shared securely.
.excalidraw file.Ensure your local environment has the required dependencies. As with other Openclaw Skills, installation is straightforward and designed to work cleanly with minimal package overhead:
# Install required client-side encryption dependencies
pip install cryptography
Once the dependencies are installed, you can generate a secure shareable link by executing the direct upload script on any local .excalidraw file:
python ~/.hermes/skills/creative/excalidraw/scripts/upload.py ~/diagrams/my_diagram.excalidraw
The diagram skill organizes its JSON configuration using standard element nodes. Every diagram file conforms to the structural schema outlined below.
| Characters in Label | Minimum Box Width | Font Size |
|---|---|---|
| 1 to 8 chars | 120 px | 16 (Virgel) |
| 9 to 14 chars | 160 px | 16 (Virgel) |
| 15 to 20 chars | 200 px | 16 (Virgel) |
| 21 to 28 chars | 240 px | 16 (Virgel) |
To prevent overflow and ensure text is perfectly legible, box dimensions are calculated dynamically using: box_width >= (text_length * 9.5) (at standard fontSize 16).
rectangle, ellipse, or diamond objects containing unique ID, position metadata, and visual configuration values.text elements bound explicitly to their parent containers via containerId and boundElements properties. Do not use raw, invalid label properties on container shapes.startBinding, endBinding) targeting specific shape handles to maintain connections even during layout updates.label attributes with exact parent-child bound elements to ensure persistent text centering and proper scaling.Loading
A data-driven cognitive toolkit to combat societal pessimism and analyze global progress using the core tenets of reason, science, humanism, and progress.

An automated operating workflow and validation rubric designed to ensure Chrome extensions meet Web Store policies, privacy promises, and high-quality release standards.

A specialized AI agent skill that communicates with the Pexo platform to draft scripts, generate scenes, and render publish-ready launch videos for startups and products.

An AI-powered video generation skill that automates scriptwriting, shot selection, voiceovers, and music production to deliver polished, startup-ready intro videos.

An advanced CLI-based browser automation skill that wraps chrome-devtools-mcp for high-performance and token-efficient AI agent web interactions.

An advanced automation skill for the SJZL platform that enables ecommerce sellers to perform one-click product listing, automatic URL identification, and configuration management.








































