Nodes are the building blocks of recipes in Neurocad. You connect them together on a canvas to define processing workflows for EDA and mechanical design tasks.
A node is a self-contained step in your recipe with specific inputs and outputs:
Inputs appear on the left side of the node
Outputs appear on the right side of the node
Each node performs a specific action: uploading a file, extracting an image, generating a 3D model, creating a library component, and so on. Nodes take input from other nodes or from direct user entry, then produce outputs that can feed into the next step.
There are a few ways to add nodes:
Browse or search the sidebar palette on the left. Nodes are grouped into categories: Input, EDA, Mechanical, and Output.
Drag a node from the palette onto the canvas to place it where you want.
Right-click anywhere on the canvas to open a context menu with all available nodes.
Drag from an open handle into empty space to get a filtered context menu showing only compatible node types.
Disabled nodes (coming soon) appear dimmed in the palette and cannot be added.
To build a workflow, connect nodes by drawing edges between their ports:
Click and drag from an output port on one node to an input port on another node.
The connection appears as a line between the two nodes.
Ports are color-coded by type, so you can visually identify which connections are valid.
The canvas enforces type compatibility. You can only connect ports that carry compatible data types (for example, text to text, image to image).
If you drag a connection to a node with multiple inputs, the canvas automatically selects the best matching port.
Nodes fall into three categories based on how they work:
Input nodes provide data to your recipe.
Add file uploads a file as a starting asset.
Prompt provides a text description to guide generation.
Extract image pulls images out of PDFs, SVGs, or 3D files.
Import brings in EDA libraries and schematics.
Corner set defines parameter variations for sweep runs.
Connect read / Connect write define the input and output boundaries of automation recipes.
Process nodes are the generative core of a recipe.
EDA process nodes live in the EDA palette section:
Library creates component symbols and footprints from reference images.
Mechanical process nodes live in the Mechanical palette section:
Model creates 3D CAD parts and enclosures from drawings.
Interpret drawing analyzes an engineering drawing and extracts structured dimensions. Its parameter output can feed directly into a Model node for more precise results.
Output nodes are the final step of a recipe. They receive the processed result and deliver it somewhere.
Live design sends the artifact to your local Neurocad installation.
Run recipe executes the full recipe as a job and can publish it as an embeddable link.
Embeddable publishes the artifact for external sharing or embedding.
Click on a node to select it. The right-side panel shows all configurable parameters for that node. Depending on the node type, you can adjust things like:
Prompt text and placeholder guidance
Image reference inputs (symbol, footprint, part, enclosure)
Operation mode and sub-action settings
Corner set parameter definitions
Ports are color-coded to help you identify compatible connections at a glance:
Text - prompts and descriptions
Image - raster images from uploads or extraction
Data - structured values like parameter sets and sweep results
EDA - electronic design artifacts (libraries, schematics, PCBs)
Mechanical - CAD artifacts (3D models, drawings, assemblies)
Generic - accepts any type, used by output nodes
Process nodes have a Run button that appears in the properties panel when the node is selected and all required inputs are connected.
You can run a single node on its own to test it, or run the full recipe end to end. Both paths use the same execution pipeline, so results are consistent whether you are testing a single step or running a complete workflow.
Last updated March 28, 2026