──────────────────────────────────────────────────────────────────
motion@0.2.1 — Parallax + TiltGallery
──────────────────────────────────────────────────────────────────
+ Parallax.tsx
- scroll-driven translate3d via rAF + window.scroll
- speed multiplier + axis (y / x) + reduced-motion bypass
- listener cleanup + cancelAnimationFrame on unmount
- WAVE 13.D.1
+ TiltGallery.tsx
- horizontally-scrolling rail of <TiltCard>-style tiles
- per-tile cursor-tracking rotateX/Y + glare gradient
- role=region + arrow-key scrolling (←/→) + scroll-snap
- reduced-motion strips tilt + glare, keeps the rail
- WAVE 13.D.5
+ 5 new tests (Parallax x 2, TiltGallery x 3) — 28/28 passing
+ index.ts: exports both + types
+ package.json: 0.2.0 → 0.2.1
──────────────────────────────────────────────────────────────────
ai-ui@0.6.0 — Wave 9.E composition surfaces
──────────────────────────────────────────────────────────────────
+ Markdown.tsx
- dep-free subset renderer: h1-h3 / **bold** / *italic* /
`code` / fenced code / ul + ol / [text](url)
- inline `[cite:<id>]` chips resolved from a citations
registry (missing ids render as [?] — failure mode is loud)
- WAVE 9.E.1
+ CodeDiff.tsx
- line-LCS diff in <100 LOC, zero deps
- split (2-col) and unified views; tinted add/del rows
- WAVE 9.E.2
+ ExplainThis.tsx
- listens for selectionchange, pops 'Explain' CTA over the
selection rect when inside the wrapper + ≥ minLength chars
- fires onExplain({ text, rect }) so hosts can open a richer
side panel if preferred
- WAVE 9.E.3
+ usePromptHistory.ts
- bash-style ↑/↓ recall with localStorage persistence
(storage key configurable; null = in-memory for tests/SSR)
- dedupes consecutive duplicates + trims to capacity
- WAVE 9.E.4
+ useTokenCount.ts
- cheap estimator (default ~4 chars/token; configurable for
code/CJK) + optional USD cost
- memoised — stable across re-renders
- WAVE 9.E.5
+ 19 new tests in src/__tests__/composition.test.tsx — 98/98 passing
+ index.ts: '0.6 surfaces' section exports all 5 + types
+ package.json: 0.5.0 → 0.6.0
Showcase routes + roadmap flips land in the paired showcase commit.
|
||
|---|---|---|
| .. | ||
| src | ||
| package.json | ||
| README.md | ||
| tsconfig.json | ||
| vitest.config.ts | ||
@bytelyst/ai-ui
AI-native UI primitives for ByteLyst products. The flagship deliverable
of Wave 2 in learning_ai_uxui_web/docs/ROADMAP_2026.md.
Status:
0.1.0MVP — 3 components + 1 hook + 1 utility. Subsequent 0.x bumps add<ToolCallCard>,<CitationChip>,<AgentTimeline>,<ModelPicker>,<ToolPalette>, and theuseToolCalls/usePromptHistory/useTokenCounthooks.
Quick start
import { ChatStream } from '@bytelyst/ai-ui';
export function ChatPage() {
return (
<div style={{ height: '70vh' }}>
<ChatStream
endpoint="/api/chat"
placeholder="Ask anything…"
emptyState={<EmptyHint />}
/>
</div>
);
}
Architectural decisions
Per learning_ai_common_plat/docs/ROADMAP_2026_DECISIONS.md:
- §10 hook shape —
useChat()returns the same shape as Vercel AI SDK'suseChat. Drop-in compatible for any engineer who has already written a chat UI with the AI SDK. - Transport is pluggable — products inject
endpoint,streamProtocol,fetcher, orheaders. We are not locked to Vercel's SSE wire format.
Components
<ChatStream>
Opinionated composition of useChat + <MessageBubble> +
<PromptComposer>. Auto-scrolls (sticky-bottom), shows Stop while
streaming, surfaces transport errors inline.
<MessageBubble>
Single chat message atom. Variants by role (user / assistant / system /
tool). Built-in "thinking" indicator while streaming with no content
yet. Honors --bl-* design tokens.
<PromptComposer>
Multi-line input. Enter submits, Shift+Enter newlines, Cmd/Ctrl+Enter
alternates submit. Auto-grows up to maxLines. Slot for leading
actions (slash-commands etc.).
Hooks
useChat(options?)
Vercel-AI-SDK-shaped:
const {
messages, input, setInput, handleInputChange, handleSubmit,
append, isLoading, error, stop, reload, setMessages
} = useChat({
endpoint: '/api/chat',
streamProtocol: 'text', // 'text' | 'sse' | 'data'
initialMessages: [],
onFinish: (msg) => console.log('done', msg),
onError: (err) => console.error(err),
});
streamText(response, protocol?, signal?)
Low-level async generator that yields text deltas from a streaming
HTTP response. Supports 'text', 'sse', and 'data' protocols.
Wire protocols
| Protocol | Body format | Use when |
|---|---|---|
'text' (default) |
Plain text, streamed chunk-by-chunk | Any backend that pipes an LLM stream directly |
'sse' |
Server-Sent Events with data: ... frames |
OpenAI-compatible, Anthropic, most HF inference |
'data' |
Vercel AI SDK "data stream" prefix protocol | Backends using streamText() from ai/server |
For tool calls / citations / images, use 'sse' or 'data' once the
matching components ship in 0.2.x.
Tests
pnpm --filter @bytelyst/ai-ui test
Roadmap
0.2.0—<ToolCallCard>,<CitationChip>,useToolCalls()0.3.0—<AgentTimeline>,<ModelPicker>0.4.0—<ToolPalette>with MCP discovery0.5.0—<Markdown>+<CodeDiff>(Wave 2 stretch)1.0.0— API freeze when 3+ products consume in production