Tekla MCP
An MCP server for Tekla that facilitates interaction with Tekla Structures, allowing users to speed-up modeling processes
What is Tekla MCP?
Tekla MCP is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to mcp server for tekla that facilitates interaction with tekla structures, allowing users to speed-up modeling processes
An MCP server for Tekla that facilitates interaction with Tekla Structures, allowing users to speed-up modeling processes
This server falls under the APIs and Cloud Services categories on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- An MCP server for Tekla that facilitates interaction with Te
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx tekla-mcp-serverConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Tekla MCP
Tekla MCP Server is a Python-based MCP server that bridges AI assistants and Tekla Structures — the industry-leading structural engineering BIM software. It exposes tools for selection, component insertion, property management, view control, drawing operations, and modeling through a natural language interface, letting engineers describe structural changes in plain text and have them executed in Tekla. The server uses a hybrid semantic system combining MiniLM embeddings and LLM-based fallback to map user-friendly descriptions to actual Tekla component properties, and supports read-only mode for safe query-only access.
Prerequisites
- Tekla Structures 2022 or 2026 installed and running on Windows (required — the server communicates with the active Tekla process)
- Python 3.11, 3.12, or 3.13 installed
- uv package manager installed (pip install uv or from astral.sh)
- An MCP-compatible client such as Claude Desktop or Cursor
- A Tekla model must be open in Tekla Structures before using MCP tools
Clone the repository and install dependencies
Clone the tekla_mcp_server repository and install all required Python packages using uv.
git clone https://github.com/teknovizier/tekla_mcp_server
cd tekla_mcp_server
uv pip install -r requirements.txtCreate and configure settings.json
Copy the sample configuration files to their active names. At minimum, configure settings.json with the correct path to your Tekla Structures binary directory.
copy config\settings.sample.json config\settings.json
copy config\element_types.sample.json config\element_types.json
copy config\semantic_overrides.sample.json config\semantic_overrides.json
copy config\base_components.sample.json config\base_components.json
copy config\report_properties.sample.json config\report_properties.jsonSet tekla_path in settings.json
Edit config/settings.json and set 'tekla_path' to the bin directory of your Tekla Structures installation. For Tekla 2022, the default is 'C:\Program Files\Tekla Structures\2022.0\bin'.
{
"tekla_path": "C:\\Program Files\\Tekla Structures\\2022.0\\bin",
"read_only": false,
"embeddings": {
"enabled": true,
"embedding_model": "teknovizier/minilm-tekla-attr-embed-v1"
}
}Configure your MCP client to launch the server
Add Tekla MCP Server to your Claude Desktop or Cursor configuration. Use absolute Windows paths and set the required environment variables.
{
"mcpServers": {
"tekla-mcp": {
"command": "python",
"args": ["C:\\path\\to\\tekla_mcp_server\\src\\tekla_mcp_server\\mcp_server.py"],
"env": {
"TEKLA_MCP_LOG_LEVEL": "INFO",
"TEKLA_MCP_LOG_FILE_PATH": "C:\\path\\to\\mcp_server.log",
"TEKLA_MCP_CONFIG_DIR": "C:\\path\\to\\tekla_mcp_server\\config"
}
}
}
}Open Tekla Structures with a model and connect your MCP client
Start Tekla Structures and open the model you want to work with. Then restart your MCP client. On first connection, the embedding model (~120 MB) will download automatically from HuggingFace.
Tekla MCP Examples
Client configuration (Claude Desktop — Windows)
Full Claude Desktop configuration for Tekla MCP Server on Windows with all required environment variables.
{
"mcpServers": {
"tekla-mcp": {
"command": "python",
"args": ["C:\\Users\\Engineer\\tekla_mcp_server\\src\\tekla_mcp_server\\mcp_server.py"],
"env": {
"TEKLA_MCP_LOG_LEVEL": "INFO",
"TEKLA_MCP_LOG_FILE_PATH": "C:\\Users\\Engineer\\tekla_mcp_server\\mcp_server.log",
"TEKLA_MCP_CONFIG_DIR": "C:\\Users\\Engineer\\tekla_mcp_server\\config"
}
}
}
}Prompts to try
Use these natural language prompts to interact with an open Tekla Structures model.
- "Select all steel beams on grid line A"
- "Insert a concrete column at coordinates X=5000, Y=3000, Z=0"
- "Show me all elements with phase number 2"
- "Change the concrete cover thickness of the selected beam to 40mm"
- "List all available components in the base components catalog"
- "Switch to the 3D view and zoom to the selected elements"Troubleshooting Tekla MCP
Server fails to connect to Tekla Structures
Ensure Tekla Structures is running and a model is open before starting or using the MCP server. Verify that 'tekla_path' in settings.json points to the correct bin directory for your Tekla version (e.g., '2022.0' or '2026.0'). For Tekla 2026, also confirm that '<tekla_path>\Net48Runtime' exists.
Python package conflict with the 'clr' module
A naming conflict can occur if the Python 'clr' package (from pythonnet) conflicts with another package. Solution: rename or delete 'C:\Users\<User>\AppData\Local\Programs\Python\Python3xx\Lib\site-packages\clr' if it belongs to a different package.
Embedding model download hangs on first startup
The MiniLM embedding model (~120 MB) downloads automatically from HuggingFace on first use. If the download hangs, manually download 'teknovizier/minilm-tekla-attr-embed-v1' from huggingface.co and set 'embedding_model' in settings.json to the local directory path. Alternatively, set 'embeddings.enabled' to false to skip semantic mapping and use LLM-only attribute resolution.
Frequently Asked Questions about Tekla MCP
What is Tekla MCP?
Tekla MCP is a Model Context Protocol (MCP) server that mcp server for tekla that facilitates interaction with tekla structures, allowing users to speed-up modeling processes It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Tekla MCP?
Follow the installation instructions on the Tekla MCP GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Tekla MCP?
Tekla MCP works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Tekla MCP free to use?
Yes, Tekla MCP is open source and available under the GPL-3.0 license. You can use it freely in both personal and commercial projects.
Tekla MCP Alternatives — Similar APIs Servers
Looking for alternatives to Tekla MCP? Here are other popular apis servers you can use with Claude, Cursor, and VS Code.
Kong
★ 43.4k🦍 The API and AI Gateway
API Mega List
★ 5.4kThis GitHub repo is a powerhouse collection of APIs you can start using immediately to build everything from simple automations to full-scale applications. One of the most valuable API lists on GitHub—period. 💪
Fetch
★ 5.4kFetch web content and convert to markdown for AI consumption
Fusio
★ 2.1kSelf-Hosted API Management for Builders
Korean Law
★ 1.8k국가법령정보MCP v4.0 | 법제처 41개 API → 17개 MCP 도구. 법령·판례·조례 검색 + LLM 환각 방지 인용검증 + 조문 영향 그래프(impact_map) + 시점 비교 자동 diff(time_travel) + 시민 5단계 실행 가이드(action_plan) | 41 Korean legal APIs → 17 MCP tools
RuleGo
★ 1.5k⛓️RuleGo is a lightweight, high-performance, embedded, next-generation component orchestration rule engine framework for Go.
Browse More APIs MCP Servers
Explore all apis servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Tekla MCP in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Tekla MCP?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.