Engineering knowledge, executable.
Genworks GDL turns design rules into live Common Lisp objects: parametric geometry, NURBS solids, web apps, and MCP tools an AI agent can operate. Lisp used to be the barrier. Now it is the leverage.
(define-object robot (base-object)
:computed-slots
((head-radius 1.0)
(body-length 3.0))
:objects
((head :type 'sphere
:radius (the head-radius))
(body :type 'box
:length (the body-length))))Why this works now
Hardcore coders get a programmable model. New KBE builders get AI assistance. Managers get durable engineering assets instead of throwaway automation scripts.
- For coders: the model is code
A GDL object is readable source, live runtime state, and design intent in one place. Edit the s-expression, query the image, regenerate the geometry. No black-box model tree.
- For new KBE builders: AI lowers the Lisp wall
GDL has always been compact and powerful; Lisp syntax kept some engineers away. With an AI pair programmer, newcomers can read, ask, generate, test, and learn against a running system.
- For managers: automation that survives
Refined since 1997, built on ANSI Common Lisp, and backed by commercial NURBS solids, non-manifold booleans, CAD exchange, and enterprise support. This is infrastructure, not a demo.
Cyclops
$99An optional but useful edge component for Genworks and skewed-emacs stacks: a small, serious reverse proxy for people who would rather configure services with symbolic expressions than wrestle YAML. Linux/x64 is available now; ARM64 and Windows are next.
Training and Certification
Learn enough GDL to build real KBE systems, with AI assistance and human instruction available when the model gets subtle.
Rent-a-Stack
Coming soonA hosted, metered Genworks GDL stack delivered over MCP—KBE modeling as a service. Higher tiers include commercial NURBS solids and non-manifold booleans. Driven by humans or autonomous agents alike. From Gornskew.
FAQ
Short answers for engineers and buyers.
How is Genworks GDL different from traditional CAD?
In Genworks GDL, your design intent lives in executable code, not frozen geometry. This means your engineering knowledge is captured as declarative definitions that update automatically when inputs change—enabling automation, optimization, and long-term flexibility that traditional CAD-only models cannot match.
Why Common Lisp?
Common Lisp combines symbolic AI heritage with industrial-strength performance. Its ANSI standard (stable since 1994) ensures your code will run for decades, while its powerful macro system and runtime flexibility make it ideal for building knowledge-based engineering systems. Franz Inc has continuously supported Allegro CL since 1984.
What is the difference between Gendl and Genworks GDL?
Gendl is the AGPL-licensed open-source kernel—perfect for research, education, and hobbyists. Genworks GDL builds on Gendl with commercial licensing, the SMLib NURBS kernel, HarmonyWare CAD I/O integration, Franz Inc partnership, and enterprise support. Both share the same core architecture refined since 1997.
How long has Genworks been in business?
Genworks International (founded as Knowledge Based Solutions in 1997) has been continuously operating for 27+ years. We've powered mission-critical applications at Fortune 500 companies since 2002 and maintain support continuity agreements with Franz Inc to ensure long-term stability.
Is this for humans or AI agents?
Both—by design, not by retrofit. Because GDL is homoiconic (code is data) and declarative (you state intent, the engine computes the result), the same define-object source a Lisp developer writes is exactly what an LLM agent emits and rewrites. The runtime is introspectable through the REPL, and the whole stack—down to the Emacs editor itself, via Elisp—is exposed over MCP. A human drives it through the editor; an agent drives it through the same tools. The symbolic foundation is what makes that confluence natural rather than forced.
Try it now
Experience Genworks GDL in Your Browser
No installation required. The same live environment can be driven by a human at the terminal or by an AI agent over MCP—same code, same tools, two operators. Explore interactive examples and see knowledge-based engineering work in real time, right in your browser.