Platform
ScaiWave ScaiGrid ScaiCore ScaiBot ScaiDrive ScaiKey Models Tools & Services
Solutions
Organisations Developers Internet Service Providers Managed Service Providers AI-in-a-Box
Resources
Support Documentation Blog Downloads
Company
About Research Careers Investment Opportunities Contact
Log in

Concepts

The mental model behind ScaiFlow.

Read these in order on first contact:

  1. Architecture — the pieces (canvas, backend, compiler, ScaiGrid, ScaiCore) and how they fit together.
  2. The flow graph — nodes, edges, ports, the JSON shape on the wire.
  3. The rigidity spectrum@rigid, @guarded, @flexible, and the "adaptive approval pattern" that composes the third with @checkpoint.
  4. Node kinds — the 24 building blocks, grouped by category.
  5. Triggers — entry nodes compile to top-level IRTrigger declarations, not to in-flow blocks. The closed set is {api, webhook, schedule}.
  6. Models and the registry — declare every model your flow needs by role; LLM nodes pick a role.
  7. Human-in-the-loop and checkpoints — pause execution, route to a review queue, resume on response.
  8. Sandboxes (ScaiBunker) — provision, exec, file I/O, destroy.
  9. Queues (ScaiQueue) — topology, publish/consume, escalation.
  10. Sub-flows — compose flows from other deployed Cores.
  11. Publish as model — turn a deployed Core into a chat model reachable over OpenAI-compatible chat completions.
  12. Multi-tenancy and auth — ScaiKey OIDC, per-tenant credentials, per-flow ACLs.
  13. Compilation targets — YAML for ScaiGrid (default), SCIR MessagePack for verification, .scai text for inspection.
Updated 2026-05-18 16:05:16 View source (.md) rev 4