Patent Diagram Generator Skill

Create patent-style technical diagrams including flowcharts, block diagrams, and system architectures using Graphviz.

When to Use

Invoke this skill when users ask to:

• Create flowcharts for method claims
• Generate block diagrams for system claims
• Draw system architecture diagrams
• Create technical illustrations for patents
• Add reference numbers to diagrams
• Generate patent figures

What This Skill Does

1. Flowchart Generation:
   • Method step flowcharts
   • Decision trees
   • Process flows with branches
   • Patent-style step numbering
2. Block Diagram Creation:
   • System component diagrams
   • Hardware architecture diagrams
   • Software module diagrams
   • Component interconnections
3. Custom Diagram Rendering:
   • Render Graphviz DOT code
   • Support multiple formats (SVG, PNG, PDF)
   • Multiple layout engines (dot, neato, fdp, circo, twopi)
4. Patent-Style Formatting:
   • Add reference numbers (10, 20, 30, etc.)
   • Use clear labels and connections
   • Professional formatting for USPTO filing

Required Dependencies

This skill requires Graphviz to be installed:

Windows:

choco install graphviz

Linux:

sudo apt install graphviz

Mac:

brew install graphviz

Python Package:

pip install graphviz

How to Use

When this skill is invoked:

1. Load diagram generator:

   import os
   import sys
   sys.path.insert(0, os.environ.get('CLAUDE_PLUGIN_ROOT', '.'))
   from mcp_server.diagram_generator import PatentDiagramGenerator
   
   generator = PatentDiagramGenerator()

2. Create flowchart from steps:

   steps = [
       {"id": "start", "label": "Start", "shape": "ellipse", "next": ["step1"]},
       {"id": "step1", "label": "Initialize System", "shape": "box", "next": ["decision"]},
       {"id": "decision", "label": "Is Valid?", "shape": "diamond", "next": ["step2", "error"]},
       {"id": "step2", "label": "Process Data", "shape": "box", "next": ["end"]},
       {"id": "error", "label": "Handle Error", "shape": "box", "next": ["end"]},
       {"id": "end", "label": "End", "shape": "ellipse", "next": []}
   ]
   
   diagram_path = generator.create_flowchart(
       steps=steps,
       filename="method_flowchart",
       output_format="svg"
   )

3. Create block diagram:

   blocks = [
       {"id": "input", "label": "Input\\nSensor", "type": "input"},
       {"id": "cpu", "label": "Central\\nProcessor", "type": "process"},
       {"id": "memory", "label": "Memory\\nStorage", "type": "storage"},
       {"id": "output", "label": "Output\\nDisplay", "type": "output"}
   ]
   
   connections = [
       ["input", "cpu", "raw data"],
       ["cpu", "memory", "store"],
       ["memory", "cpu", "retrieve"],
       ["cpu", "output", "processed data"]
   ]
   
   diagram_path = generator.create_block_diagram(
       blocks=blocks,
       connections=connections,
       filename="system_diagram",
       output_format="svg"
   )

4. Render custom DOT code:

   dot_code = """
   digraph PatentSystem {
       rankdir=LR;
       node [shape=box, style=rounded];
   
       Input [label="User Input\\n(10)"];
       Processor [label="Processing Unit\\n(20)"];
       Output [label="Display\\n(30)"];
   
       Input -> Processor [label="data"];
       Processor -> Output [label="result"];
   }
   """
   
   diagram_path = generator.render_dot_diagram(
       dot_code=dot_code,
       filename="custom_diagram",
       output_format="svg",
       engine="dot"
   )

5. Add reference numbers:

   # After creating a diagram, add patent-style reference numbers
   reference_map = {
       "Input Sensor": 10,
       "Central Processor": 20,
       "Memory Storage": 30,
       "Output Display": 40
   }
   
   annotated_path = generator.add_reference_numbers(
       svg_path=diagram_path,
       reference_map=reference_map
   )

Diagram Templates

Get common templates:

templates = generator.get_diagram_templates()

# Available templates:
# - simple_flowchart: Basic process flow
# - system_block: System architecture
# - method_steps: Sequential method
# - component_hierarchy: Hierarchical structure

Shape Types

Flowchart Shapes

• ellipse: Start/End points
• box: Process steps
• diamond: Decision points
• parallelogram: Input/Output operations
• cylinder: Database/Storage

Block Diagram Types

• input: Input devices/sensors
• output: Output devices/displays
• process: Processing units
• storage: Memory/storage
• decision: Control logic
• default: General components

Layout Engines

• dot: Hierarchical (top-down/left-right)
• neato: Spring model layout
• fdp: Force-directed layout
• circo: Circular layout
• twopi: Radial layout

Output Formats

• svg: Scalable Vector Graphics (best for editing)
• png: Raster image (good for viewing)
• pdf: Portable Document Format (USPTO compatible)

Patent-Style Reference Numbers

Convention:

• Main components: 10, 20, 30, 40, ...
• Sub-components: 12, 14, 16 (under 10)
• Elements: 22, 24, 26 (under 20)

Example labeling:

"Input Sensor (10)"
"  - Detector Element (12)"
"  - Signal Processor (14)"
"Central Unit (20)"
"  - CPU Core (22)"
"  - Cache (24)"

Presentation Format

When creating diagrams:

1. Describe what will be generated: "Creating a flowchart for the authentication method with 5 steps..."
2. Generate the diagram: Run Python code to create SVG/PNG/PDF
3. Show file location: "Diagram created: ${CLAUDE_PLUGIN_ROOT}/python\diagrams\method_flowchart.svg"
4. List reference numbers (if added):

   Reference Numbers:
   - Input Module (10)
   - Processing Unit (20)
   - Output Interface (30)

Common Use Cases

1. Method Claims → Flowcharts
   • Show sequential steps
   • Include decision branches
   • Number steps (S1, S2, S3...)
2. System Claims → Block Diagrams
   • Show components and connections
   • Use reference numbers
   • Indicate data flow directions
3. Architecture Diagrams → Custom DOT
   • Complex system layouts
   • Multiple interconnections
   • Hierarchical structures

Error Handling

If Graphviz is not installed:

1. Check installation: dot -V
2. Install for your OS (see above)
3. Verify Python package: pip show graphviz
4. Test generation: python scripts/test_diagrams.py

Tools Available

• Bash: To run Python diagram generation
• Write: To save DOT code or diagrams
• Read: To load existing diagrams or templates