OmniGraffle BindCraft
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@OmniGraffle BindCraft Create a flowchart for user login process"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
omnigraffle-bindcraft
OmniGraffle automation skill + MCP server for AI coding agents. Create publication-quality scientific diagrams, flowcharts, and architecture diagrams via natural language.
Features
36 MCP tools for full OmniGraffle control: shapes, connections, layout, export
Smart arrow routing — auto-selects straight / orthogonal / curved based on shape positions; hop-over arcs at crossings
6 journal color palettes — Nature, Science, Cell, Lancet, Minimal, Vibrant
6 diagram templates — experimental workflow, data pipeline, PRISMA, system architecture, ML pipeline, clinical trial
2658 IconPark icons across 38 categories
Batch operations — create dozens of shapes and connections in single calls
Hub magnets — distribute connection points to prevent arrow pile-ups
Dual automation — Omni Automation JS + JXA/AppleScript
Related MCP server: drawio-mcp-server
Requirements
macOS (OmniGraffle is macOS-only)
OmniGraffle 7 (Pro recommended for full scripting support)
Node.js >= 18
Python 3 with
cairosvg— for icon SVG-to-PNG conversion (optional, only needed for icons)
Installation
Quick install (auto-detects runtime)
git clone https://github.com/Youn-17/omnigraffle-bindcraft.git
cd omnigraffle-bindcraft
npm install
bash scripts/install.shThe install script auto-detects Claude Code, Codex, and WorkBuddy, then:
Builds the project
Adds the MCP server to each runtime's config
Copies the skill file to each runtime's skills directory
Manual install
npm run buildAdd to ~/.claude/settings.json:
{
"mcpServers": {
"omnigraffle": {
"command": "node",
"args": ["/absolute/path/to/omnigraffle-bindcraft/dist/index.js"]
}
}
}Install the skill:
cp .claude/skills/sci-diagram.md ~/.claude/skills/omnigraffle-bindcraft.mdnpm run buildAdd to ~/.workbuddy/.mcp.json:
{
"mcpServers": {
"omnigraffle": {
"command": "node",
"args": ["/absolute/path/to/omnigraffle-bindcraft/dist/index.js"]
}
}
}Install the skill:
cp .claude/skills/sci-diagram.md ~/.workbuddy/skills/omnigraffle-bindcraft.mdnpm run buildAdd to ~/.codex/config.json:
{
"mcpServers": {
"omnigraffle": {
"command": "node",
"args": ["/absolute/path/to/omnigraffle-bindcraft/dist/index.js"]
}
}
}Any agent supporting the Model Context Protocol can use this server. Add the MCP entry to your agent's config — the format is the same across runtimes.
Set up IconPark icons (optional)
git clone https://github.com/bytedance/IconPark.git resources/iconpark
pip install cairosvgQuick start
Once installed, just describe what you want in natural language:
Create a RAG pipeline diagram in OmniGraffle with 10 components:
User Query, Query Analyzer, Intent Router, Knowledge Retrieval,
LLM Reasoning Core, Tool Executor, Evidence Checker,
Response Generator, Safety Filter, Final Response.
Use Nature palette. Export to Desktop.See examples/demo_rag_pipeline.md for a full walkthrough.
Tools (36)
Document & Canvas (8)
Tool | Description |
| Get document name, path, canvases |
| Create new document (optionally from template) |
| Save current document |
| List all available OmniGraffle templates |
| Create a new canvas |
| List all canvases |
| Switch to a canvas by index |
| Set canvas dimensions |
Shapes (6)
Tool | Description |
| Create a shape with type, position, size, text, and full styling |
| Create multiple shapes in one call |
| Update any property of an existing graphic |
| Delete graphics by ID |
| List all graphics with properties |
| Remove all graphics from canvas |
Connections (4)
Tool | Description |
| Connect two shapes with styled, auto-routed line |
| Batch connect with smart routing ( |
| Create a standalone line between points |
| Set connection points on a shape for clean arrow distribution |
Layout & Organization (6)
Tool | Description |
| Apply automatic layout |
| Align graphics (left/center/right/top/bottom) |
| Create a new layer |
| List all layers |
| Group graphics together |
| Ungroup a group |
Icons (5)
Tool | Description |
| Search 2658 IconPark icons by keyword |
| List all 38 icon categories |
| Get icon suggestions for a scientific context |
| Insert a colored icon onto the canvas |
| Create a labeled shape with an icon inside |
Palettes & Templates (4)
Tool | Description |
| List all 6 scientific color palettes |
| Get a palette's full color values |
| List all 6 diagram templates |
| Get a template's nodes, icons, and palette |
Export & Scripting (4)
Tool | Description |
| Export to PNG, PDF, or SVG |
| Execute Omni Automation JavaScript |
| Execute JXA code |
| Execute AppleScript |
Smart arrow routing
The connect_shapes_batch tool supports line_type: "auto" which automatically picks the best routing:
Shape relationship | Auto-selected type | Why |
Same column (Δx < 60px) |
| Clean vertical drop |
Same row (Δy < 30px) |
| Direct horizontal |
Cross-column & cross-row |
| Right-angle turns |
Target above source |
| Avoids overlap with forward flow |
All connections automatically get HopType.Round for clean crossing arcs.
Color palettes
Palette | Best for | Style |
| General science papers | Clean, high-contrast, colorblind-safe |
| Bold research graphics | Vivid but balanced |
| Biology, life sciences | Elegant muted tones |
| Clinical, medical | Professional ocean blue + orange |
| Grayscale print | Maximum print compatibility |
| Posters, presentations | High-saturation, eye-catching |
Architecture
AI Agent <── MCP Protocol ──> MCP Server (Node.js)
│
┌─────────────┼─────────────┐
│ │ │
Omni Automation JXA/AS IconPark
(evaluateJS) (osascript) (cairosvg)
│ │ │
└──────┬──────┘ SVG → PNG
│
OmniGraffle 7Development
npm run dev # Watch mode — recompiles on changes
npm run build # One-time build
npm start # Run the server directlyLicense
MIT
Credits
IconPark by ByteDance — 2658 high-quality SVG icons
OmniGraffle by The Omni Group
Model Context Protocol by Anthropic
Available Tools
37 toolsalign_graphicsC
Align multiple graphics relative to each other
Inputs: ids (required), alignment (required).
auto_layoutA
Apply automatic layout to the current canvas. OmniGraffle will arrange all connected shapes.
Inputs: no named parameters.
clear_canvasA
Remove all graphics from the current canvas
Inputs: no named parameters.
connect_shapesB
Connect two shapes with a line/arrow. The line automatically routes between the shapes.
Inputs: source_id (required), target_id (required), label, head_type, +4 more.
connect_shapes_batchA
Connect multiple pairs of shapes at once with full styling. Supports auto-routing: when line_type is 'auto', picks straight for vertical/horizontal neighbors, orthogonal for cross-column, curved for feedback loops.
Inputs: connections (required).
create_canvasB
Create a new canvas in the current document
Inputs: name (required).
create_documentA
Create a new OmniGraffle document, optionally from a built-in template
Inputs: template.
create_layerB
Create a new layer on the current canvas
Inputs: name (required), visible.
create_lineB
Create a standalone line between two points (not attached to shapes)
Inputs: points (required), label, head_type, tail_type, +2 more.
create_node_with_iconB
Create a labeled shape with an icon inside — perfect for scientific diagram nodes. The icon is placed above the text label within the shape.
Inputs: x (required), y (required), label (required), icon_name (required), +9 more.
create_shapeC
Create a shape on the current canvas with position, size, text, and styling
Inputs: x (required), y (required), width (required), height (required), +12 more.
create_shapes_batchA
Create multiple shapes at once. Returns an array of created shape IDs. More efficient than calling create_shape multiple times.
Inputs: shapes (required).
delete_graphicsB
Delete one or more graphics by ID
Inputs: ids (required).
export_canvasB
Export the current canvas to an image file (PNG, PDF, or SVG)
Inputs: format (required), output_path (required), scope.
get_diagram_templateA
Get full details of a diagram template including nodes, suggested icons, and palette
Inputs: template (required).
get_document_infoA
Get information about the current OmniGraffle document: name, canvases, current canvas, and graphics count
Inputs: no named parameters.
get_graphicsA
List all graphics on the current canvas with their properties (ID, type, position, size, text, style)
Inputs: canvas_index.
get_paletteB
Get a specific color palette with all color values ready to use
Inputs: palette (required).
group_graphicsC
Group multiple graphics together
Inputs: ids (required).
insert_iconA
Insert an IconPark icon onto the OmniGraffle canvas as an image. The icon is rendered as a PNG with optional color customization.
Inputs: x (required), y (required), icon_name (required), size, +2 more.
list_canvasesA
List all canvases in the current document with their graphics count
Inputs: no named parameters.
list_diagram_templatesB
List available scientific diagram templates with their node structures and suggested palettes/icons
Inputs: no named parameters.
list_icon_categoriesA
List all IconPark icon categories with their icon counts
Inputs: no named parameters.
list_layersA
List all layers on the current canvas
Inputs: no named parameters.
list_palettesA
List available scientific color palettes (Nature, Science, Cell, Lancet, Minimal, Vibrant)
Inputs: no named parameters.
list_templatesA
List all available OmniGraffle templates
Inputs: no named parameters.
run_applescriptA
Execute AppleScript code. Use for operations that work better with AppleScript syntax.
Inputs: code (required).
run_jxaA
Execute JXA (JavaScript for Automation) code via osascript. Use for operations that need the macOS scripting bridge, such as file system access combined with OmniGraffle control.
Inputs: code (required).
run_omni_jsA
Execute arbitrary Omni Automation JavaScript inside OmniGraffle. The code runs in OmniGraffle's JS context with access to the full Omni Automation API. Use for operations not covered by other tools.
Inputs: code (required).
save_documentA
Save the current OmniGraffle document
Inputs: no named parameters.
search_iconsA
Search IconPark icons by keyword. Returns matching icon names that can be used with insert_icon. 2658 icons available across 38 categories.
Inputs: query (required), limit.
set_canvas_sizeC
Set the size of the current canvas
Inputs: width (required), height (required), canvas_index.
set_magnetsA
Set magnet (connection) points on a shape. Magnets define where lines can attach.
Inputs: id (required), magnets (required).
suggest_iconsB
Get icon suggestions for a scientific diagram context. Provide a description and get relevant icon names.
Inputs: context (required).
switch_canvasB
Switch to a different canvas by index (0-based)
Inputs: index (required).
ungroupB
Ungroup a group graphic
Inputs: id (required).
update_graphicC
Update properties of an existing graphic by ID
Inputs: id (required), x, y, name, +12 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 37 tools
Most tools have clearly distinct purposes, but some overlap exists (e.g., connect_shapes vs connect_shapes_batch, run_* tools) that could cause minor confusion. Descriptions help differentiate, but an agent might still select the wrong tool occasionally.
The majority of tools follow verb_noun pattern (create_shape, list_canvases), but a few deviate (auto_layout, ungroup) and run_applescript/run_jxa use different casing. Overall consistent and readable.
37 tools is on the higher side, covering many features of OmniGraffle and including auxiliary tools (icon search, palettes). While each serves a purpose, the set feels somewhat heavy and could be streamlined by removing rarely-used scripting tools.
The tool set covers core CRUD operations, layout, grouping, exporting, and includes batch operations and scripting. Minor gaps like missing import or full shape modification properties, but generally well-rounded for diagram creation and automation.
Related MCP Connectors
Draw architecture diagrams, UML, planning boards and clickable wireframes from your AI assistant.
1Create and edit architecture diagrams from your AI agent; get an SVG and a live editable canvas.
Create, edit and organize hand-drawn diagrams (flowcharts, ER, architecture) from any AI agent.
Turn calls and notes into process flow diagrams in Claude, ChatGPT, VS Code, Cursor, and Teams.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceEnables AI agents to create visual, interactive interfaces with diagrams, structured input forms, and persistent state management instead of text-only responses.7-
- AlicenseNot gradedqualityDmaintenanceEnables AI agents to create, read, update, and delete Draw.io diagrams, allowing automated generation of architectural diagrams, flowcharts, and visual documentation.MIT
- AlicenseAqualityDmaintenanceEnables AI agents to generate and edit draw.io diagrams with real-time browser preview.559 npmApache 2.0
- FlicenseNot gradedqualityAmaintenanceEnables AI coding assistants to turn natural-language descriptions of systems and data into editable, auto-laid-out Excalidraw diagrams across 25 types — from architecture and sequence diagrams to Gantt charts and scatter plots — plus optionally persist the architecture as a queryable, version-controlled knowledge graph. Runs fully offline with export to SVG and PNG, and diagrams can be revised afterward in natural language.-