Skip to main contentBuilding blocks
- Input Node — Named entry ports.
- Processing Nodes — Actions/Agents/Flows.
- Output Node — Named exit ports.
- Edges — Data/control movement between Nodes.
Patterns
- Linear — Input → Action → Output.
- Branching — Route by condition into multiple paths.
- Parallel — Fan‑out work; aggregate results downstream.
- Iterative — Loop over collections or conditional repetition using nested loop structures.
Enhanced Flow System
Recent improvements to the Flow system include:
- Better validation — Enhanced graph validation catches structural errors early
- Loop support — Native support for nested flow structures with iteration
- Improved schema handling — Better validation logic for complex data flows
- Build/edit operations — Expanded flow interface allows iterative development
Practices
- Keep Flows legible top down.
- Favor shallow graphs; compose via sub‑Flows for depth.
- Remove unused ports and edges to reduce confusion.
- Use loops for batch processing and iterative operations.
- Validate flow structure early to catch connection issues.
tip: Design for testing — Save example Payloads at the Flow level to lock in expected shapes and make regression checks easy.
