Development platform for .NET-based CAD/BIM applications — currently supporting Autodesk Revit (2022–2027) and AutoCAD-family (2022–2027), designed to extend to any .NET-capable host.

Essential shortcuts for rapid development:

• Press AD — re-run last script + show Trace Panel
• Ctrl + Click Trace Panel button — toggle panel visibility
• F5 in VSCode — attach debugger for breakpoint debugging

Basic workflow:

1. Click script in Code Execution panel → First execution
2. Edit code in your IDE
3. Press AD → See updated results
4. Repeat steps 2-3 for rapid iteration

Debugging workflow:

1. Set breakpoints in VSCode
2. Press F5 → Attach to host Python
3. Execute script in RevitDevTool
4. Debugger pauses at breakpoints
5. Inspect variables, step through code

Access RevitDevTool settings from the dockable panel to configure themes, logging, visualization, and code execution behavior:

Multi-language code execution framework

• Execute Python 3.13 scripts with CPython + PythonNet3
• Automatic dependency management with PEP 723 + Pixi
• VSCode debugger integration with breakpoints and variable inspection
• .NET hot-reload with FileWatcher
• C# / F# script execution via Roslyn
• Hierarchical tree organization

Use when: You need to run custom scripts or commands with automatic dependency installation.

Learn more: Execution Overview | Modern Python Scripting | Python Debugging

Model Context Protocol for AI-powered workflows

• Standalone MCPServer.exe discovers all running host instances
• Built-in tools: list_host_instances, launch_host, open_model, execute_csharp_code, open_document
• Custom Python/C# toolsets via folder convention
• Works with Cursor, Claude Desktop, VS Code Copilot, any MCP client

Use when: You want AI assistants to interact with live Revit/AutoCAD instances.

Learn more: AI Tool Integration

Run tests inside live host processes

• pytest plugin runs tests inside Revit, AutoCAD, Civil 3D
• Auto-discovery of running host instances via Named Pipe
• CLI: --host revit, --host autocad, --host civil3d
• IDE integration (VS Code, Cursor, PyCharm)

Use when: You need automated API testing against a live host.

Learn more: PyTest Bridge

Unified logging infrastructure

• Multi-sink output (UI RichTextBox + File logging)
• Host context enrichment (version, document, user info)
• Syntax highlighting with keyword detection
• Pretty JSON output
• Python stack trace formatting
• Geometry interception — automatic visualization (Revit)

Use when: You need structured logging with visual feedback.

Learn more: Logging Overview | Color Keywords

Real-time 3D geometry rendering

• DirectContext3D transient rendering (no model elements)
• Curves, faces, solids, meshes, points, bounding boxes
• Thread-safe buffering with performance optimization

Use when: You need to visualize geometry without creating model elements.

Learn more: Visualization Overview | Geometry Types

flowchart TD
    UI[User Interaction<br/>Execute script, visualize geometry, view logs]

    UI --> CE[Code Execution<br/>Discovers scripts and assemblies<br/>Manages dependencies via Pixi<br/>Executes code in host context]

    CE --> Output[Code execution output]

    Output --> Logging[Logging<br/>Captures output<br/>Applies themes<br/>Routes to sinks]

    Output --> Viz[Visualization<br/>Renders 3D<br/>DirectContext3D<br/>Revit only]

    Logging <-->|Geometry?| Viz

    Logging --> TracePanel[Trace Panel<br/>with colors]
    Viz --> View3D[3D View<br/>with geometry]

    MCP[MCP Server<br/>AI assistants<br/>Multi-host discovery]
    MCP -->|tools/prompts| CE

    PyTest[PyTest Bridge<br/>Remote test execution<br/>Multi-host]
    PyTest -->|tests/run| CE

Python script with code execution + logging + visualization:

# /// script
# dependencies = ["polars==1.38.1"]
# ///

import polars as pl
from Autodesk.Revit.DB import FilteredElementCollector, Wall

doc = __revit__.ActiveUIDocument.Document
walls = FilteredElementCollector(doc).OfClass(Wall).ToElements()

print(f"Found {len(list(walls))} walls")

for wall in walls:
    curve = wall.Location.Curve
    print(curve)  # Geometry automatically visualized

print("Analysis complete")

What happens:

1. Code Execution discovers script, installs polars via Pixi, executes code
2. Logging captures print() calls — Trace panel with syntax highlighting
3. Logging intercepts print(geometry) — routes to Visualization
4. Visualization renders curves in 3D via DirectContext3D
5. User sees both text output and visual output

More examples: Examples Overview

• Data Analysis — Collect, analyze, visualize
• Curve Visualization — Pick and display curves
• Dashboard — WebView2 dashboard with charts
• Civil 3D Examples — Alignments, surfaces, corridors

Learn more: Host Support | Revit Features | AutoCAD Features | Civil 3D Features

RevitDevTool works with multiple programming languages:

• Python 3.13 — full support via CPython + PythonNet3
• C# — full support through .NET and Roslyn scripts
• F# — script execution support
• IronPython — first-class script mode with pyRevit runtime support
• Any .NET Language — VB.NET, etc. via assembly execution

1. Download the latest release from GitHub Releases
2. Run the MSI installer
3. Launch Revit (2022–2027) or AutoCAD (2022–2027)
4. Open the RevitDevTool panel from the ribbon

For developers contributing to RevitDevTool:

• Architecture Overview
• Execution Architecture
• MCP Architecture
• PyTest Architecture
• Logging Architecture
• Visualization Architecture

• Main Repository: RevitDevTool
• Issues: Report bugs or request features
• Discussions: Ask questions

MIT License — See LICENSE in main repository.