freecad-chatcad

AI agent side panel for FreeCAD: model parametric parts by chatting with Claude.

PluginOpen sourceMaintainedMCAD

What it does and how it connects

Chat CAD for FreeCAD is an AI agent side panel docked inside FreeCAD. Describe a part, point at a face, or scribble on the 3D view, and an agent models it as a parametric FreeCAD document you can keep tweaking by hand. It works with Claude Code, OpenAI Codex, Google Antigravity, or a local model through Ollama or any OpenAI-compatible server. Dimensions live in a Parametros spreadsheet with aliases, so changing a number and pressing Ctrl+R updates the part. Status is alpha, tested on Windows 11 with FreeCAD 1.1.4; Linux and macOS should work but are untested, and the interface is in Brazilian Portuguese.

Safety nets include a backup copy before changes, one undo entry per local-agent step, and automated checks that verify each result. The README is candid about local models: in its tests every real multi-feature part failed with local models, even one feature at a time, while Codex was the fastest and most faithful cloud agent in a timed comparison, taking 6 min 21 s with a volume within 0.04 percent of the reference part.

Agents run Python code inside the user's FreeCAD session with the user's permissions. CLI agents are sandboxed: Codex runs with a read-only sandbox, Claude Code loads only the freecad MCP server, and Antigravity allows only mcp(freecad/*). The panel talks to FreeCAD through the freecad-mcp server by neka-nat.

Recurring errors it addresses

Error Index entries this tool is built to reduce or work around.

Sources

  1. 1READMEgithub.com
  2. 2GitHub repo statsapi.github.com

Not verified

  • Pricing: none stated on the repo.
  • Real user voices: no real linked post read for this page; the repo has no public issues.
  • No GitHub releases exist, so latest release fields are null.
  • A standalone website: the repo homepage points to the author's portfolio, not a product site.

Last verified November 6, 2026. Listings are informational. The evidence score measures public, verifiable evidence and is not an endorsement or a compatibility claim.