AI agent UI runtime — wire up LLM agents with server-streamed Geometra UIs

Streamed React Pixi.jscomponent

Experimental credential-free Codex request generation and streamed response parsing for Loom.js.

Returns Protobuf schemas for the data streamed and stored in S3 for a given topic

Dynamic memory for AI teams — one living, verified memory, streamed to every agent. Thin client for the Warmboot relay.

Node/browser SDK for emitting **Avada product behavior events** to the ingest pipeline (`POST /ingest/events`). Events are typed against the central event registry, batched client-side, and streamed into `avada_product_raw.{appId}_sdk_events`.

Streamed Web Server with easy Ajax, EventSource, Templating, Static pages; Middleware approach and extensibility.

Smoothly reveal streamed (LLM) text in React — backend-agnostic, zero-dependency.

Provider-neutral streamed model and tool orchestration protocol for Loom.js coding agents.

Transform incomplete markdown into stable, parseable output to prevent visual glitches in AI streamed content.

Three.js render target helpers for Geometra — stack WebGL with the geometry-streamed canvas client, declarative scene3d element rendering

Transform stream which splits streamed data.

Headless Vue components for building TanStack AI chat interfaces with streamed message parts.

Headless Solid components for building TanStack AI chat interfaces with streamed message parts.

Transform stream for inspecting streamed data.

Browser client for viewing pixel-streamed content from MSquared events

EarthOS streamed imagery tiles: web-mercator quadtree draped on the globe

The browser runtime for LiteShip: apply CSS, streamed HTML, worker output, and LLM chunks to a live DOM with focus- and scroll-preserving morphing.

Text-Streamed React component

Private in-browser translation with streamed previews and single-transaction Yjs and BlockNote commits.