PlantUML CSA MCP Server
Generate Control System Architecture (CSA) diagrams using PlantUML via the Model Context Protocol (MCP). Supports ISA-95 Purdue model, industrial symbols, and multiple protocols.
README
PlantUML CSA MCP Server
Generate Control System Architecture (CSA) diagrams using PlantUML via the Model Context Protocol (MCP).
Features
- ISA-95 Purdue Model - Zone packages with color-coded levels (0-4)
- Industrial Symbols - 36+ component types (10 controllers, 26 devices)
- Protocol Visualization - 12 protocols with distinct line colors
- Architecture Templates - 5 pre-defined templates for common configurations
- Bootstrap from Skills - Generate topology from equipment-list + instrument-io skills
- Multiple Layout Engines - Graphviz, Smetana, ELK
- Shareable URLs - Generate PlantUML server links
- Same Schema as FreeCAD - Interoperable YAML topology format
Project Structure
plantuml-csa-mcp-server/
├── src/
│ ├── __init__.py # Package entry point
│ ├── __main__.py # Module runner (python -m src)
│ ├── server.py # FastMCP server with 8 MCP tools
│ ├── models/
│ │ └── csa_topology.py # Pydantic models (same as FreeCAD CSA)
│ ├── converter/
│ │ ├── topology_to_puml.py # YAML topology → PlantUML source
│ │ ├── sprites.py # ISA symbol definitions (36+ types)
│ │ └── layout_hints.py # Purdue-aware layout directives
│ ├── renderer/
│ │ └── plantuml_runner.py # PlantUML CLI wrapper (JAR/native/Docker)
│ ├── encoder/
│ │ └── plantuml_encoder.py # Text encoding for shareable URLs
│ ├── templates/
│ │ └── architecture_templates.py # 5 architecture templates
│ └── bootstrap/
│ └── csa_bootstrap.py # Bootstrap from equipment/IO lists
├── tests/ # 56 tests
├── docs/
│ └── completed-plans/ # Implementation plan
├── pyproject.toml
├── CLAUDE.md # Development guide
└── README.md
Installation
# Clone and install
cd plantuml-csa-mcp-server
uv sync
# Verify PlantUML is available
uv run python -c "from src.renderer import PlantUMLRunner; print(PlantUMLRunner().check_available())"
Quick Start
1. Add to MCP Configuration
Add to your .mcp.json:
{
"mcpServers": {
"plantuml-csa-mcp": {
"type": "stdio",
"command": "uv",
"args": ["--directory", "/path/to/plantuml-csa-mcp-server", "run", "python", "-m", "src"],
"description": "PlantUML CSA diagram generation (ISA-95 Purdue model)"
}
}
}
2. Generate a Diagram
# topology.yaml
schema_version: "1.0"
metadata:
project_name: "Sample WWTP CSA"
zones:
- id: "level_0"
purdue_level: 0
- id: "level_1"
purdue_level: 1
controllers:
- id: "PLC-101"
type: PLC
zone: "level_1"
devices:
- id: "RIO-101"
type: RemoteIO
parent_controller: "PLC-101"
zone: "level_0"
links:
- source: "PLC-101"
target: "RIO-101"
protocol: "Ethernet_IP"
Then use the MCP tools:
# Get PlantUML source for version control
csa_get_plantuml_source(topology_yaml=yaml_content)
# Generate SVG diagram
csa_generate_diagram(topology_yaml=yaml_content, format="svg")
# Get shareable URL
csa_encode_plantuml(plantuml_source=puml_source)
MCP Tools Reference
| Tool | Purpose |
|---|---|
csa_generate_diagram |
Render topology YAML to SVG/PNG |
csa_get_plantuml_source |
Get .puml source for version control |
csa_validate_topology |
Validate YAML against schema |
csa_list_symbols |
List available ISA-style symbols |
csa_encode_plantuml |
Generate shareable PlantUML URLs |
csa_list_templates |
List architecture templates |
csa_render_preview |
Quick preview for iteration |
csa_check_plantuml |
Check PlantUML availability |
csa_bootstrap_from_io |
Bootstrap topology from equipment/IO lists |
Architecture Templates
Pre-defined templates for common control system configurations:
| Template | Description | Use Case |
|---|---|---|
centralized |
Central MCC + Central PLC | Small plants (<5 MLD) |
central_mcc_distributed_io |
Central MCC + Distributed IO | Medium plants (5-20 MLD) |
fully_distributed |
Remote panels per area | Large plants (>20 MLD) |
hybrid_safety |
Central Safety + Distributed Process | SIL/SIS requirements |
vendor_package_integration |
OEM packages via OPC-UA | Multiple vendor packages |
Bootstrap from Skill Outputs
Generate CSA topology from equipment-list-skill and instrument-io-skill outputs:
csa_bootstrap_from_io(
equipment_list_qmd=equipment_qmd_content,
instrument_database_yaml=io_database_yaml,
project_name="WWTP Control System",
architecture_template="fully_distributed"
)
# Returns: {topology_yaml, suggestions, io_summary, equipment_mapping}
Supported Components
Controllers (10 types)
PLC, DCS, PAC, Safety_PLC, Soft_PLC, Edge_Controller, Motion_Controller, Redundant_PLC, RTU, SIS
Devices (26 types)
RemoteIO, HMI, SCADA, Historian, OPC_UA_Server, Gateway, VFD, Soft_Starter, MCC, Industrial_PC, Switch, Managed_Switch, Router, Firewall, Wireless_AP, Media_Converter, Network_TAP, Motor_Starter, Engineering_WS, Panel_PC, Data_Logger, Junction_Box, Marshalling_Cabinet, Local_Panel, Remote_Panel, Instrument_Rack
Protocols (12 types)
Ethernet_IP, Profinet, Modbus_TCP, Modbus_RTU, Profibus, DeviceNet, ControlNet, HART, Foundation_Fieldbus, OPC_UA, MQTT, BACnet
Testing
# Run all tests
uv run pytest tests/ -v
# Run with coverage
uv run pytest tests/ --cov=src
vs FreeCAD CSA
| Aspect | PlantUML CSA | FreeCAD CSA |
|---|---|---|
| Output | SVG/PNG | TechDraw PDF |
| Version Control | Plain text .puml | Binary .FCStd |
| Dependencies | Java/PlantUML | FreeCAD runtime |
| Best For | Documentation | CAD deliverables |
Use Both: Same YAML topology works with both renderers. Use PlantUML for rapid iteration and documentation, FreeCAD for final engineering drawings.
Workflow Integration
This server is part of the puran-water control system architecture workflow:
┌─────────────────────────┐ ┌──────────────────────────┐ ┌─────────────────────┐
│ equipment-list-skill │ ──► │ instrument-io-skill │ ──► │ csa-diagram-skill │
│ (equipment + feeder) │ │ (IO lists, patterns) │ │ (CSA generation) │
└─────────────────────────┘ └──────────────────────────┘ └─────────────────────┘
│
▼
┌─────────────────────┐
│ plantuml-csa-mcp │
│ (this server) │
└─────────────────────┘
│
▼
┌─────────────────────┐
│ CSA Topology YAML │
│ PlantUML PNG/SVG │
│ Shareable URLs │
└─────────────────────┘
Related Projects
Upstream (Data Sources)
- equipment-list-skill - Equipment lists with feeder types
- instrument-io-skill - IO lists with DI/DO/AI/AO patterns
Companion Skill
- csa-diagram-skill - Claude Code skill that orchestrates this MCP server
Similar Pattern (Electrical)
- plantuml-sld-mcp-server - Single-Line Diagram generation (same YAML → PlantUML pattern)
- electrical-distribution-skill - SLD skill using plantuml-sld-mcp-server
Alternative Renderer
- freecad-csa-workbench - Same YAML topology rendered in FreeCAD for CAD deliverables
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.