Revit MCP
Revit-MCP
English |简体中文
描述
revit-mcp 允许您通过 MCP 支持的客户端(例如 Claude、Cline 等)使用 MCP 协议与 Revit 进行交互。
本项目是服务器端(提供Tools给AI),需要配合revit-mcp-plugin (驱动Revit)使用。
Related MCP server: RevitMCPBridge2026
特征
允许 AI 从 Revit 项目获取数据
允许AI驱动Revit创建、修改和删除元素
将AI生成的代码发送给Revit执行(不一定成功,在一些简单且需求明确的场景下成功率更高)
要求
Node.js 18+
完整安装环境仍需考虑revit-mcp-plugin的需求,请参考revit-mcp-plugin
安装
1. 搭建本地MCP服务
安装依赖项
npm install建造
npm run build2.客户端配置
克劳德客户
Claude 客户端 -> 设置 > 开发者 > 编辑配置 > claude_desktop_config.json
{
"mcpServers": {
"revit-mcp": {
"command": "node",
"args": ["<path to the built file>\\build\\index.js"]
}
}
}重启 Claude 客户端,当看到锤子图标时,表示 MCP 服务连接正常。

框架
flowchart LR
CladueDesktop --> revit-mcp --> SocketService--commandName-->CommandlSet--command-->CommandExecute
CommandManager --> CommandlSet
CommandExecute --executeResult--> SocketService
CommandProject1 --> CommandManager
CommandProject2 --> CommandManager
CommandProject... --> CommandManager
subgraph ide1 [MCPClient]
CladueDesktop
end
subgraph ide2 [MCPServer]
revit-mcp
end
subgraph ide3 [Revit]
subgraph ide3.1 [revit-mcp-plugin]
SocketService
CommandlSet
CommandManager
CommandExecute
end
end支持的工具
姓名 | 描述 |
获取当前视图信息 | 获取当前视图信息 |
获取当前视图元素 | 获取当前视图元素 |
获取可用的家庭类型 | 获取当前项目中可用的族类型 |
获取选定元素 | 获取选定元素 |
创建基于点的元素 | 创建基于点的元素(门、窗、家具) |
创建基于线的元素 | 创建基于线的元素(墙、梁、管道) |
创建基于表面的元素 | 创建基于表面的元素(地板、天花板) |
删除元素 | 删除元素 |
重置模型 | 重置模型(执行连续对话时删除流程模型) |
修改元素 | 修改元素的属性(实例参数) |
搜索模块 | 搜索可用模块 |
use_module | 使用模块 |
发送代码到 Revit | 将代码发送到 Revit 执行 |
颜色飞溅 | 根据参数值对元素进行着色 |
标签墙 | 标记视图中的所有墙壁 |
Available Tools
13 toolsai_element_filterA
An intelligent Revit element querying tool designed specifically for AI assistants to retrieve detailed element information from Revit projects. This tool allows the AI to request elements matching specific criteria (such as category, type, visibility, or spatial location) and then perform further analysis on the returned data to answer complex user queries about Revit model elements. Example: When a user asks 'Find all walls taller than 5m in the project', the AI would: 1) Call this tool with parameters: {"filterCategory": "OST_Walls", "includeInstances": true}, 2) Receive detailed information about all wall instances in the project, 3) Process the returned data to filter walls with height > 5000mm, 4) Present the filtered results to the user with relevant details.
Inputs: data (required).
color_elementsA
Color elements in the current view based on a category and parameter value. Each unique parameter value gets assigned a distinct color.
Inputs: categoryName (required), parameterName (required), useGradient, customColors.
create_line_based_elementA
Create one or more line-based elements in Revit such as walls, beams, or pipes. Supports batch creation with detailed parameters including family type ID, start and end points, thickness, height, and level information. All units are in millimeters (mm).
Inputs: data (required).
create_point_based_elementA
Create one or more point-based elements in Revit such as doors, windows, or furniture. Supports batch creation with detailed parameters including family type ID, position, dimensions, and level information. All units are in millimeters (mm).
Inputs: data (required).
create_surface_based_elementA
Create one or more surface-based elements in Revit such as floors, ceilings, or roofs. Supports batch creation with detailed parameters including family type ID, boundary lines, thickness, and level information. All units are in millimeters (mm).
Inputs: data (required).
delete_elementA
Delete one or more elements from the Revit model by their element IDs.
Inputs: elementIds (required).
get_available_family_typesA
Get available family types in the current Revit project. You can filter by category and family name, and limit the number of returned types.
Inputs: categoryList, familyNameFilter, limit.
get_current_view_elementsA
Get elements from the current active view in Revit. You can filter by model categories (like Walls, Floors) or annotation categories (like Dimensions, Text). Use includeHidden to show/hide invisible elements and limit to control the number of returned elements.
Inputs: modelCategoryList, annotationCategoryList, includeHidden, limit.
get_current_view_infoA
获取 Revit 当前活动视图的详细信息,包括视图类型、名称、比例等属性。
Inputs: no named parameters.
get_selected_elementsB
Get elements currently selected in Revit. You can limit the number of returned elements.
Inputs: limit.
operate_elementC
Operate on Revit elements by performing actions such as select, selectionBox, setColor, setTransparency, delete, hide, etc.
Inputs: data (required).
send_code_to_revitB
Send C# code to Revit for execution. The code will be inserted into a template with access to the Revit Document and parameters. Your code should be written to work within the Execute method of the template.
Inputs: code (required), parameters.
tag_all_wallsA
Create tags for all walls in the current active view. Tags will be placed at the middle point of each wall.
Inputs: useLeader, tagTypeId.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 13 tools
Most tools have distinct purposes, such as ai_element_filter for querying, create_*_element for element creation by type, and delete_element for removal. However, there is some overlap between ai_element_filter, get_current_view_elements, and get_selected_elements, as all retrieve elements but with different scopes (filtered vs. view vs. selected), which could cause confusion if not carefully described. The descriptions help clarify, but the boundaries are not entirely clear-cut.
The naming follows a consistent verb_noun pattern for most tools, such as create_line_based_element, delete_element, and get_available_family_types. However, there are minor deviations: ai_element_filter uses a less standard prefix, operate_element is vague compared to others, and tag_all_walls is more specific in naming. Overall, the conventions are readable and mostly uniform, with only slight inconsistencies.
With 13 tools, the count is well-scoped for a Revit MCP server covering element creation, querying, deletion, and operations. Each tool appears to serve a distinct function in the BIM workflow, such as creating different element types, filtering, and executing code, making the set comprehensive without being overwhelming. This aligns with typical domain-specific servers that require moderate tool counts for effective coverage.
The tool set provides strong coverage for core Revit operations, including creation (point, line, surface-based), deletion, querying (filtering, view info, selection), and specialized actions like tagging and coding. Minor gaps exist, such as lack of update or modify tools for existing elements beyond basic operations in operate_element, and no explicit tools for managing views or parameters beyond current view info. However, agents can likely work around these with the available tools, ensuring most workflows are supported.
Maintenance
Related MCP Connectors
Read and edit the Revit 2024–2027 model on your PC. Needs the free Revit MCP add-in.
Global CAD/BIM, industrial engineering (FEM, CFD) and property tools via one MCP endpoint.
Convert Revit files to XKT, IFC, or DWG and query BIM data via natural language.
AI for Revit: 450+ tools, 1 undo step. Windows, Revit 2024–2027 (2021–2023 included). Free account.
Related MCP Servers
- AlicenseBqualityNot gradedmaintenanceAllows AI assistants to interact with Autodesk Revit through the MCP protocol, enabling the AI to create, modify, and delete elements in Revit projects.1559 npm1-
- AlicenseNot gradedqualityAmaintenanceMCP server for Autodesk Revit (BIM) with 705+ API endpoints. Enables AI agents to create walls, place doors/windows, generate sheets, manage views, and produce construction documents via the Model Context Protocol. Uses named pipes for zero-crash Revit integration.22MIT
- AlicenseAqualityFmaintenanceEnables AI to interact with Revit via MCP, allowing data retrieval and element creation, modification, and deletion.13MIT
- AlicenseNot gradedqualityCmaintenanceEnables AI agents to connect to Revit 2024 via MCP, inspect documents and selections, query and modify elements, and validate, preview, or commit plans with audit logging and rollback.MIT