Start from a working step,
not a blank canvas.
A useful workflow contains steps worth repeating: extracting a record, checking a requirement, preparing a decision. Compose these into reusable networks with explicit inputs and outputs, then adapt the connections and rules to the next process.
Illustrated workflow composition. See the recorded demo for actual application footage.
See what can come next.
Ready
Starts here
Not yet
Steps that connect by what they carry.
A support workflow can combine a ticket parser, a source-grounded answer, an escalation decision and an engineering handoff. Configure each interface and test the resulting chain against routine and exceptional tickets.
Written once. Only the values change.
The card
The form
The graph
// Card "name": "Extract Action Items", "short": "Pull every commitment out of a transcript with its owner, due date and the quote that proves it.", "in": [{ "label": "transcript", "type": "text" }], "out": [ { "label": "action_items", "type": "json" }, { "label": "action_table", "type": "text" } ], // Form "variables": [{ "name": "context", "kind": "text", "required": false, "hint": "Anything the transcript assumes: the team, the project, the quarter…" }], // Graph — excerpt net.Conversation("extract").label("Extract Action Items") .schema("ActionItems") .system("You extract action items from a meeting transcript. … {{context|raw}}")
Use AI to help build. Inspect what it changes.
Set a goal
Choose a component
Correct it in plain words
Validate and run
Illustrative component configuration. Review AI-generated connections, code and values before running the workflow.
Where a process holds its standard.
Revise until it passes.
Check before posting.
Review the supporting evidence.
Build, change and run with AI.
In NodeFox
From your own agent
A reusable workflow retains the integration logic, checks and execution structure across runs. Measure model usage, elapsed time and review effort for your process rather than assuming a universal cost or latency advantage.
Give reusable logic an owner.
A copied graph and a referenced sub-network have different maintenance implications. Choose deliberately, document the dependency and validate changes before rollout.
Questions about patterns
Are patterns templates?
If a pattern changes, do my networks change?
Which models do patterns use?
Does AI decide what runs?
Which agents work over MCP?
Do patterns guarantee good output?
Can we write our own?
Build one step you can reuse.
Choose a recurring check or transformation. Define its contract, test it and reuse it in the next workflow.