Search & the node palette
Two fast ways to move: press Tab to create a node, press Cmd-F to find one you already have.
DNA graphs grow quickly. Instead of hunting through menus, you do almost everything from the keyboard — type a few letters and the right node appears. There are two tools that look similar but do different jobs: the node palette (for adding nodes) and search (for jumping to existing nodes).
The node palette — press Tab
Hover over an empty spot in the The node graph and press Tab. A search box pops up under your cursor. Start typing and DNA fuzzy-matches as you go — you don't need the exact name.
Matching is forgiving. Typing circ finds shape.circle, iter finds utility.iterator, and blur finds the blur node even though "blur" isn't the start of its full title. It also matches on what a node does, so a rough guess at a tag or category usually lands you in the right neighbourhood.
Press Enter (or click) to drop the node onto the canvas. If you opened the palette while dragging out of an existing node's output, the new node wires itself in automatically.
Pull a wire out from a node's output into empty space and let go — the palette opens already filtered to nodes that fit, and connects the result for you.
Filtering by family and tag
When you're browsing rather than searching, the palette groups everything into families so you can narrow down fast:
Generate — nodes that make new content (shapes, geometry, sources)
Modify — nodes that change what's coming in
Data — numbers, signals, attributes
IO — bringing things in and sending them out
Structure — organising the graph itself
Plugin — anything added by a plugin in
~/.dna/plugins/
Pick a family to show only those nodes. You can also filter by tags — finer labels like "2d", "audio", or "noise". Tags stack with AND logic: turn on two tags and you only see nodes that carry both. Only families and tags that actually exist in your project show up, so the lists stay tidy.
Finding a node you already have — press Cmd-F
Cmd-F opens the find box. This searches the nodes already in your graph rather than creating new ones — handy when a project has grown large and you've lost track of "that colour node I set up earlier".
Type a fuzzy query and DNA ranks the matches. It looks at more than just titles: it also searches each node's type, its category, its tags, and even its parameter names and values. So you can find "the node with a radius of 200" or "the one tagged audio" without remembering what you called it.
Use the arrow keys to move through results (the list wraps around), press Enter to jump straight to the selected node, and Escape to close. While the find box is open, nodes that don't match dim down so the matches stand out on the canvas.
Search reaches inside Subgraphs too. If a match lives in a nested subgraph, the result shows a breadcrumb path and DNA switches into that subgraph when you jump to it. Matches in your current graph are always ranked above deeper ones.
Which one do I want?
Adding something new? Tab.
Going to something that already exists? Cmd-F.
Both share the same friendly fuzzy matching, so the muscle memory carries over — the only thing to remember is create versus find.