Concepts
KDCube AI App is a self-hosted, multi-tenant platform and SDK for building AI assistants and agentic applications. The platform has two core services: Ingress (API gateway, auth, SSE emitter) and Processor/Proc (bundle execution, queue worker, integrations API).
System at a Glance
Key Terms
| Term | Description |
|---|---|
| Bundle | Python package registered with @agentic_workflow. The unit of AI application deployment — contains agent logic, tools, skills, and UI widgets. |
| Ingress | API gateway service: auth, SSE streaming, task enqueueing, rate limiting. Handles all inbound traffic before the bundle sees it. |
| Processor / Proc | Queue worker: executes bundles, hosts the Operations REST API. Loads bundle singletons and calls execute_core() per turn. |
| Timeline | Rolling cross-conversation context persisted as artifact:conv.timeline.v1. Ordered blocks (oldest → newest) streamed as SSE events. Grows across turns; compacted when context budget is reached. The turn log is the per-turn portion appended each turn. |
| ReAct Agent | Autonomous loop (ReactStateV2, max 15 iterations by default). Plan is a tool — not a separate component. No coordinator needed. |
| Tenant / Project | Multi-tenancy units. All data, config, bundles, and budgets are scoped to tenant + project. Each tenant gets an isolated schema in PostgreSQL. |
Supported Providers
Transports: SSE · Socket.IO · REST |
LLMs: OpenAI · Anthropic · Gemini · OpenRouter |
Search: Brave · DuckDuckGo |
Auth: Cognito · SimpleIDP · Delegated