diff --git a/templates/assistant-ui/.cursor/rules/echo_rules.mdc b/templates/assistant-ui/.cursor/rules/echo_rules.mdc
new file mode 100644
index 000000000..81a5c4a2e
--- /dev/null
+++ b/templates/assistant-ui/.cursor/rules/echo_rules.mdc
@@ -0,0 +1,52 @@
+---
+description: Guidelines for building Echo-powered chat applications with assistant-ui, AI SDK v5, and the assistant-ui React component library
+globs: **/*.ts,**/*.tsx,**/*.js,**/*.jsx
+---
+
+# Echo Assistant-UI Guidelines
+
+## SDK Setup
+- Server-side: Initialize in `src/echo/index.ts` with `@merit-systems/echo-next-sdk`
+- Client-side: Wrap app with `EchoProvider` from `@merit-systems/echo-next-sdk/client`
+
+## Assistant-UI Integration
+
+### Runtime Provider
+```typescript
+'use client';
+import { AssistantRuntimeProvider, useEdgeRuntime } from '@assistant-ui/react';
+
+export function AssistantProvider({ children }: { children: React.ReactNode }) {
+  const runtime = useEdgeRuntime({ api: '/api/chat' });
+  return (
+    <AssistantRuntimeProvider runtime={runtime}>
+      {children}
+    </AssistantRuntimeProvider>
+  );
+}
+```
+
+### Chat Components
+- Use `@assistant-ui/react` primitives for chat UI:
+  - `Thread` — main chat container
+  - `ThreadWelcome` — welcome screen
+  - `Composer` — message input
+  - `AssistantMessage` / `UserMessage` — message bubbles
+- Use `@assistant-ui/react-markdown` for rendering markdown in messages
+- Use `@assistant-ui/react-ai-sdk` for AI SDK v5 integration
+
+### Streaming API Route
+```typescript
+import { streamText } from 'ai';
+import { openai } from '@/echo';
+
+export async function POST(req: Request) {
+  const { messages } = await req.json();
+  const result = streamText({ model: openai('gpt-4o'), messages });
+  return result.toDataStreamResponse();
+}
+```
+
+## Environment Variables
+- `ECHO_APP_ID` — Server-side Echo identifier
+- `NEXT_PUBLIC_ECHO_APP_ID` — Client-side Echo identifier
diff --git a/templates/authjs/.cursor/rules/echo_rules.mdc b/templates/authjs/.cursor/rules/echo_rules.mdc
new file mode 100644
index 000000000..e242f21e0
--- /dev/null
+++ b/templates/authjs/.cursor/rules/echo_rules.mdc
@@ -0,0 +1,51 @@
+---
+description: Guidelines for building Echo-powered Next.js applications with Auth.js (NextAuth v5) authentication, including provider setup and session management
+globs: **/*.ts,**/*.tsx,**/*.js,**/*.jsx
+---
+
+# Echo Auth.js Integration Guidelines
+
+## SDK Setup
+- Server-side: Initialize Echo in `src/echo/index.ts` with `@merit-systems/echo-next-sdk`
+- Auth: Configure Auth.js with the Echo provider
+
+## Auth.js Configuration
+
+### Auth Setup
+```typescript
+import NextAuth from 'next-auth';
+import EchoProvider from '@merit-systems/echo-authjs-provider';
+
+export const { handlers, signIn, signOut, auth } = NextAuth({
+  providers: [
+    EchoProvider({
+      clientId: process.env.ECHO_APP_ID!,
+      // Additional provider options
+    }),
+  ],
+});
+```
+
+### API Route
+- Mount Auth.js handlers at `src/app/api/auth/[...nextauth]/route.ts`:
+```typescript
+import { handlers } from '@/auth';
+export const { GET, POST } = handlers;
+```
+
+### Session Access
+- Server components: `const session = await auth()`
+- Client components: Use `useSession()` from `next-auth/react`
+- Protect routes with middleware or `auth()` checks
+
+## Echo TypeScript SDK
+- Use `@merit-systems/echo-typescript-sdk` for server-side Echo API calls:
+```typescript
+import { EchoClient } from '@merit-systems/echo-typescript-sdk';
+```
+- Combine with `auth()` session to make authenticated Echo API requests
+
+## Environment Variables
+- `ECHO_APP_ID` — Echo application identifier (also used as Auth.js client ID)
+- `AUTH_SECRET` — NextAuth secret for session encryption
+- `NEXTAUTH_URL` — Deployment URL for Auth.js callbacks
diff --git a/templates/echo-cli/.cursor/rules/echo_rules.mdc b/templates/echo-cli/.cursor/rules/echo_rules.mdc
new file mode 100644
index 000000000..f9348371c
--- /dev/null
+++ b/templates/echo-cli/.cursor/rules/echo_rules.mdc
@@ -0,0 +1,98 @@
+---
+description: Guidelines and best practices for building Echo-powered CLI applications, including SDK setup, authentication flows, wallet integration, and AI chat sessions
+globs: **/*.ts,**/*.js
+---
+
+# Echo CLI Guidelines
+
+## SDK Setup
+
+### TypeScript SDK
+- Use `@merit-systems/echo-typescript-sdk` for server/CLI contexts:
+```typescript
+import { EchoClient } from '@merit-systems/echo-typescript-sdk';
+```
+- Use `@merit-systems/ai-x402` for x402 payment protocol integration
+
+### AI SDK Integration
+- Import AI models from `@ai-sdk/openai` for CLI chat:
+```typescript
+import { openai } from '@ai-sdk/openai';
+import { streamText } from 'ai';
+
+const result = streamText({
+  model: openai('gpt-4o'),
+  messages,
+});
+```
+
+## Authentication
+
+### Auth Methods
+The CLI supports two authentication flows:
+1. **Echo OAuth** — Browser-based OAuth with PKCE
+2. **Crypto Wallet** — WalletConnect or local wallet with private key
+
+```typescript
+import { loginWithEcho, loginWithWallet, initLocalWallet } from '@/auth';
+```
+
+### Session Management
+- Use `conf` package for persistent configuration storage
+- Check `isAuthenticated()` before making API calls
+- Use `logout()` to clear stored credentials
+
+## CLI Framework
+
+### Commander.js
+- Define commands with `commander`:
+```typescript
+import { Command } from 'commander';
+
+const program = new Command();
+program
+  .name('echodex')
+  .description('CLI Coding Agent Powered by Echo')
+  .version('1.0.0');
+
+program.command('chat').description('Start AI chat').action(startChatSession);
+```
+
+### Interactive Prompts
+- Use `@clack/prompts` for interactive selection:
+```typescript
+import { select, isCancel } from '@clack/prompts';
+
+const choice = await select({
+  message: 'Choose an option:',
+  options: [{ value: 'chat', label: 'Chat' }],
+});
+if (isCancel(choice)) process.exit(0);
+```
+
+### Output Formatting
+- Use `chalk` for colored terminal output
+- Use `ora` for spinners during async operations
+- Use `cli-table3` for tabular data display
+
+## Environment Variables
+- `ECHO_APP_ID` — Echo application identifier
+- Store in `.env.local` file
+- Load with `dotenv` or `tsx --env-file=.env.local`
+
+## Project Structure
+```
+src/
+  index.ts            # CLI entry point (commander setup)
+  auth/               # Authentication flows
+  core/               # Chat sessions, model selection, profile
+  utils/              # Helper functions
+  constants/          # Static values, ASCII art
+  print/              # Formatted output helpers
+```
+
+## Build & Development
+- Dev: `tsx --env-file=.env.local src/index.ts`
+- Build: `node scripts/build.js` (outputs to `dist/`)
+- Type check: `tsc --noEmit`
+- Package manager: pnpm
diff --git a/templates/next-chat/.cursor/rules/echo_rules.mdc b/templates/next-chat/.cursor/rules/echo_rules.mdc
new file mode 100644
index 000000000..204530960
--- /dev/null
+++ b/templates/next-chat/.cursor/rules/echo_rules.mdc
@@ -0,0 +1,56 @@
+---
+description: Guidelines for building Echo-powered Next.js chat applications with AI SDK streaming, shadcn/ui components, and real-time message handling
+globs: **/*.ts,**/*.tsx,**/*.js,**/*.jsx
+---
+
+# Echo Next.js Chat Guidelines
+
+## SDK Setup
+- Server-side: Initialize in `src/echo/index.ts` with `@merit-systems/echo-next-sdk`
+- Client-side: Wrap app with `EchoProvider` from `@merit-systems/echo-next-sdk/client`
+- Use `EchoTokens` component for authentication and token management
+
+## Chat Integration
+
+### AI SDK useChat Hook
+```typescript
+'use client';
+import { useChat } from '@ai-sdk/react';
+
+export function Chat() {
+  const { messages, input, handleInputChange, handleSubmit, isLoading } = useChat({
+    api: '/api/chat',
+  });
+  return (
+    <form onSubmit={handleSubmit}>
+      {messages.map((m) => (
+        <div key={m.id}>{m.role}: {m.content}</div>
+      ))}
+      <input value={input} onChange={handleInputChange} />
+    </form>
+  );
+}
+```
+
+### Streaming API Route
+```typescript
+import { streamText } from 'ai';
+import { openai } from '@/echo';
+import { isSignedIn } from '@/echo';
+
+export async function POST(req: Request) {
+  await isSignedIn();
+  const { messages } = await req.json();
+  const result = streamText({ model: openai('gpt-4o'), messages });
+  return result.toDataStreamResponse();
+}
+```
+
+## UI Components
+- Use shadcn/ui components (`class-variance-authority` for variants)
+- Import from `@/components/ui/` — button, input, card, scroll-area
+- Use Tailwind CSS v4 for styling
+
+## Environment Variables
+- `ECHO_APP_ID` — Server-side Echo identifier
+- `NEXT_PUBLIC_ECHO_APP_ID` — Client-side Echo identifier
diff --git a/templates/next-image/.cursor/rules/echo_rules.mdc b/templates/next-image/.cursor/rules/echo_rules.mdc
new file mode 100644
index 000000000..58af93462
--- /dev/null
+++ b/templates/next-image/.cursor/rules/echo_rules.mdc
@@ -0,0 +1,44 @@
+---
+description: Guidelines for building Echo-powered Next.js image generation applications with AI SDK, image models, and gallery components
+globs: **/*.ts,**/*.tsx,**/*.js,**/*.jsx
+---
+
+# Echo Next.js Image Generation Guidelines
+
+## SDK Setup
+- Server-side: Initialize in `src/echo/index.ts` with `@merit-systems/echo-next-sdk`
+- Client-side: Wrap app with `EchoProvider` from `@merit-systems/echo-next-sdk/client`
+- Use `EchoTokens` component for authentication and token management
+
+## Image Generation
+
+### API Route for Image Generation
+```typescript
+import { openai } from '@/echo';
+import { isSignedIn } from '@/echo';
+import { experimental_generateImage as generateImage } from 'ai';
+
+export async function POST(req: Request) {
+  await isSignedIn();
+  const { prompt } = await req.json();
+  const { image } = await generateImage({
+    model: openai.image('dall-e-3'),
+    prompt,
+  });
+  return Response.json({ url: image.base64 });
+}
+```
+
+### Client-Side Image Display
+- Display generated images with proper loading states
+- Use `next/image` for optimized image rendering when possible
+- Store image URLs/base64 in component state
+
+## UI Components
+- Use shadcn/ui components (`class-variance-authority` for variants)
+- Build a gallery component for displaying generated images
+- Include prompt input, generate button, and loading spinner
+
+## Environment Variables
+- `ECHO_APP_ID` — Server-side Echo identifier
+- `NEXT_PUBLIC_ECHO_APP_ID` — Client-side Echo identifier
diff --git a/templates/next-video-template/.cursor/rules/echo_rules.mdc b/templates/next-video-template/.cursor/rules/echo_rules.mdc
new file mode 100644
index 000000000..ab5222374
--- /dev/null
+++ b/templates/next-video-template/.cursor/rules/echo_rules.mdc
@@ -0,0 +1,41 @@
+---
+description: Guidelines for building Echo-powered Next.js video generation applications with AI video models and video player components
+globs: **/*.ts,**/*.tsx,**/*.js,**/*.jsx
+---
+
+# Echo Next.js Video Generation Guidelines
+
+## SDK Setup
+- Server-side: Initialize in `src/echo/index.ts` with `@merit-systems/echo-next-sdk`
+- Client-side: Wrap app with `EchoProvider` from `@merit-systems/echo-next-sdk/client`
+- Use `EchoTokens` component for authentication and token management
+
+## Video Generation
+
+### API Route for Video Generation
+```typescript
+import { openai } from '@/echo';
+import { isSignedIn } from '@/echo';
+
+export async function POST(req: Request) {
+  await isSignedIn();
+  const { prompt } = await req.json();
+  // Use the Echo-proxied model for video generation
+  // The specific model depends on your Echo app configuration
+  return Response.json({ videoUrl: result.url });
+}
+```
+
+### Video Player Component
+- Build a custom video player or use HTML5 `<video>` element
+- Handle video loading states and error fallbacks
+- Support playback controls: play, pause, seek, volume
+
+## UI Components
+- Use shadcn/ui components (`class-variance-authority` for variants)
+- Build a video gallery with thumbnails
+- Include prompt input, generate button, and progress indicator
+
+## Environment Variables
+- `ECHO_APP_ID` — Server-side Echo identifier
+- `NEXT_PUBLIC_ECHO_APP_ID` — Client-side Echo identifier
diff --git a/templates/next/.cursor/rules/echo_rules.mdc b/templates/next/.cursor/rules/echo_rules.mdc
new file mode 100644
index 000000000..aa3eeb9c7
--- /dev/null
+++ b/templates/next/.cursor/rules/echo_rules.mdc
@@ -0,0 +1,108 @@
+---
+description: Guidelines and best practices for building Echo-powered Next.js applications, including SDK setup, provider configuration, API routes, and AI integration
+globs: **/*.ts,**/*.tsx,**/*.js,**/*.jsx
+---
+
+# Echo Next.js Guidelines
+
+## SDK Setup
+
+### Server-Side Initialization
+- Initialize the Echo SDK in a dedicated `src/echo/index.ts` file:
+```typescript
+import Echo from '@merit-systems/echo-next-sdk';
+
+export const { handlers, isSignedIn, openai, anthropic } = Echo({
+  appId: process.env.ECHO_APP_ID!,
+});
+```
+- `ECHO_APP_ID` must be set in your `.env.local` file (server-side, no `NEXT_PUBLIC_` prefix)
+- `NEXT_PUBLIC_ECHO_APP_ID` is used only in the client-side `EchoProvider`
+
+### Client-Side Provider
+- Wrap your app with `EchoProvider` in a client component:
+```typescript
+'use client';
+import { EchoProvider } from '@merit-systems/echo-next-sdk/client';
+
+export function Providers({ children }: { children: React.ReactNode }) {
+  return (
+    <EchoProvider config={{ appId: process.env.NEXT_PUBLIC_ECHO_APP_ID! }}>
+      {children}
+    </EchoProvider>
+  );
+}
+```
+- Always add `'use client'` directive to components using `EchoProvider` or `EchoTokens`
+
+### API Route Handler
+- Create a catch-all API route at `src/app/api/echo/[...echo]/route.ts`:
+```typescript
+import { handlers } from '@/echo';
+export const { GET, POST } = handlers;
+```
+- This route handles Echo authentication and token management automatically
+
+## AI Integration
+
+### Using AI SDK with Echo
+- Always import AI models from the Echo SDK, not directly from `@ai-sdk/openai`:
+```typescript
+// ✅ Correct — uses Echo billing
+import { openai } from '@/echo';
+
+// ❌ Wrong — bypasses Echo billing
+import { openai } from '@ai-sdk/openai';
+```
+- Use Vercel AI SDK v5 patterns for streaming:
+```typescript
+import { streamText } from 'ai';
+import { openai } from '@/echo';
+
+const result = streamText({
+  model: openai('gpt-4o'),
+  messages,
+});
+```
+
+### Auth Check
+- Use `isSignedIn()` from `@/echo` to verify authentication in API routes before making AI calls
+
+## Components
+
+### EchoTokens
+- Use `EchoTokens` component for the authentication/token purchase UI:
+```typescript
+'use client';
+import { EchoTokens } from '@merit-systems/echo-next-sdk/client';
+export function EchoButton() {
+  return <EchoTokens />;
+}
+```
+- Must be in a client component (`'use client'`)
+
+## Environment Variables
+- `ECHO_APP_ID` — Server-side Echo app identifier
+- `NEXT_PUBLIC_ECHO_APP_ID` — Client-side Echo app identifier (same value)
+- Never commit `.env.local` to version control
+
+## Project Structure
+```
+src/
+  echo/
+    index.ts          # Echo SDK init (server)
+  providers.tsx       # EchoProvider wrapper (client)
+  app/
+    layout.tsx        # Root layout with Providers
+    page.tsx          # App pages
+    api/echo/[...echo]/
+      route.ts        # Echo API handler
+  components/
+    EchoWrapper.tsx   # Client-side Echo components
+```
+
+## Common Patterns
+- Always use Next.js 15 App Router (not Pages Router)
+- Use `--turbopack` flag for dev server: `next dev --turbopack`
+- Use Tailwind CSS v4 for styling
+- Use `@/` path alias for imports from `src/`
diff --git a/templates/nextjs-api-key-template/.cursor/rules/echo_rules.mdc b/templates/nextjs-api-key-template/.cursor/rules/echo_rules.mdc
new file mode 100644
index 000000000..0f9017b18
--- /dev/null
+++ b/templates/nextjs-api-key-template/.cursor/rules/echo_rules.mdc
@@ -0,0 +1,62 @@
+---
+description: Guidelines for building Echo-powered Next.js applications with server-side API key management, PostgreSQL database, and Prisma ORM
+globs: **/*.ts,**/*.tsx,**/*.js,**/*.jsx,**/*.prisma
+---
+
+# Echo Next.js API Key Template Guidelines
+
+## SDK Setup
+- Server-side: Initialize Echo in `src/echo/index.ts` with `@merit-systems/echo-next-sdk`
+- Client-side: Wrap app with `EchoProvider` from `@merit-systems/echo-next-sdk/client`
+- Also uses `@merit-systems/echo-typescript-sdk` for server-side API calls
+
+## Database (Prisma + PostgreSQL)
+
+### Schema
+- Define models in `prisma/schema.prisma`
+- Use `@prisma/client` for database queries
+- PostgreSQL as the database provider
+
+### Prisma Commands
+```bash
+npx prisma generate     # Generate client from schema
+npx prisma db push      # Push schema to database
+npx prisma migrate dev  # Create and apply migrations
+npx prisma studio       # Open database browser
+```
+
+### Client Instantiation
+```typescript
+import { PrismaClient } from '@prisma/client';
+
+const globalForPrisma = globalThis as unknown as { prisma: PrismaClient };
+export const prisma = globalForPrisma.prisma ?? new PrismaClient();
+if (process.env.NODE_ENV !== 'production') globalForPrisma.prisma = prisma;
+```
+- Use the global singleton pattern to prevent too many connections in development
+
+## API Key Management
+- Store API keys securely in the database (hashed)
+- Validate API keys on each request in middleware or API routes
+- Rate limit API key usage
+- Support key rotation and revocation
+
+## AI Integration
+```typescript
+import { openai } from '@/echo';
+import { streamText } from 'ai';
+import { isSignedIn } from '@/echo';
+```
+- Always verify authentication before processing AI requests
+- Use Echo-proxied models to enable user billing
+
+## Environment Variables
+- `ECHO_APP_ID` — Server-side Echo app identifier
+- `NEXT_PUBLIC_ECHO_APP_ID` — Client-side Echo app identifier
+- `DATABASE_URL` — PostgreSQL connection string
+- `DIRECT_URL` — Direct database URL (if using connection pooling)
+
+## UI Components
+- Use shadcn/ui components (`class-variance-authority` for variants)
+- Tailwind CSS for styling
+- Build API key management dashboard components
diff --git a/templates/react-chat/.cursor/rules/echo_rules.mdc b/templates/react-chat/.cursor/rules/echo_rules.mdc
new file mode 100644
index 000000000..b30eff733
--- /dev/null
+++ b/templates/react-chat/.cursor/rules/echo_rules.mdc
@@ -0,0 +1,45 @@
+---
+description: Guidelines for building Echo-powered React chat applications with Vite, AI SDK streaming, and shadcn/ui components
+globs: **/*.ts,**/*.tsx,**/*.js,**/*.jsx
+---
+
+# Echo React Chat Guidelines
+
+## SDK Setup
+- Wrap app with `EchoProvider` from `@merit-systems/echo-react-sdk`
+- Use `import.meta.env.VITE_ECHO_APP_ID` for the Echo app identifier
+- Use `EchoTokens` component for authentication and token management
+
+## Chat Integration
+
+### AI SDK useChat Hook
+```typescript
+import { useChat } from '@ai-sdk/react';
+
+export function Chat() {
+  const { messages, input, handleInputChange, handleSubmit, isLoading } = useChat({
+    api: '/api/chat',
+  });
+  return (
+    <form onSubmit={handleSubmit}>
+      {messages.map((m) => (
+        <div key={m.id}>{m.role}: {m.content}</div>
+      ))}
+      <input value={input} onChange={handleInputChange} />
+    </form>
+  );
+}
+```
+
+## UI Components
+- Use shadcn/ui components (`class-variance-authority` for variants)
+- Tailwind CSS for styling
+- Build message bubbles, input area, and sidebar components
+
+## Environment Variables
+- `VITE_ECHO_APP_ID` — Client-side Echo identifier (Vite prefix required)
+
+## Build & Development
+- Dev: `npm run dev` (Vite)
+- Build: `tsc -b && vite build`
+- This is a client-side only app — no SSR
diff --git a/templates/react-image/.cursor/rules/echo_rules.mdc b/templates/react-image/.cursor/rules/echo_rules.mdc
new file mode 100644
index 000000000..7284ca4bd
--- /dev/null
+++ b/templates/react-image/.cursor/rules/echo_rules.mdc
@@ -0,0 +1,29 @@
+---
+description: Guidelines for building Echo-powered React image generation applications with Vite and AI image models
+globs: **/*.ts,**/*.tsx,**/*.js,**/*.jsx
+---
+
+# Echo React Image Generation Guidelines
+
+## SDK Setup
+- Wrap app with `EchoProvider` from `@merit-systems/echo-react-sdk`
+- Use `import.meta.env.VITE_ECHO_APP_ID` for the Echo app identifier
+- Use `EchoTokens` component for authentication and token management
+
+## Image Generation
+- Use `@ai-sdk/react` hooks for prompt submission
+- Display generated images with loading states
+- Build a gallery component for browsing generated images
+
+## UI Components
+- Use shadcn/ui components (`class-variance-authority` for variants)
+- Tailwind CSS for styling
+- Include prompt input, generate button, and image grid
+
+## Environment Variables
+- `VITE_ECHO_APP_ID` — Client-side Echo identifier (Vite prefix required)
+
+## Build & Development
+- Dev: `npm run dev` (Vite)
+- Build: `tsc -b && vite build`
+- This is a client-side only app — no SSR
diff --git a/templates/react/.cursor/rules/echo_rules.mdc b/templates/react/.cursor/rules/echo_rules.mdc
new file mode 100644
index 000000000..3a574778b
--- /dev/null
+++ b/templates/react/.cursor/rules/echo_rules.mdc
@@ -0,0 +1,75 @@
+---
+description: Guidelines and best practices for building Echo-powered React applications with Vite, including SDK setup, provider configuration, and AI integration
+globs: **/*.ts,**/*.tsx,**/*.js,**/*.jsx
+---
+
+# Echo React (Vite) Guidelines
+
+## SDK Setup
+
+### Provider Configuration
+- Wrap your app with `EchoProvider` in `src/main.tsx`:
+```typescript
+import { EchoProvider } from '@merit-systems/echo-react-sdk';
+
+ReactDOM.createRoot(document.getElementById('root')!).render(
+  <EchoProvider config={{ appId: import.meta.env.VITE_ECHO_APP_ID }}>
+    <App />
+  </EchoProvider>
+);
+```
+- Use `import.meta.env.VITE_ECHO_APP_ID` for environment variables (Vite convention)
+- All client-exposed env vars must be prefixed with `VITE_`
+
+### Authentication UI
+- Use the `EchoTokens` component for user auth and token purchases:
+```typescript
+import { EchoTokens } from '@merit-systems/echo-react-sdk';
+
+function App() {
+  return <EchoTokens />;
+}
+```
+
+## AI Integration
+
+### Using AI SDK v5 with Echo
+- Use `@ai-sdk/react` hooks for chat and completion UIs:
+```typescript
+import { useChat } from '@ai-sdk/react';
+
+function ChatComponent() {
+  const { messages, input, handleInputChange, handleSubmit } = useChat({
+    api: '/api/chat',
+  });
+}
+```
+- AI SDK v5 is pinned — always use the version specified in `package.json`
+- Use `overrides` in `package.json` to prevent version conflicts between SDK packages
+
+### React SDK Imports
+```typescript
+// ✅ Correct — client-side imports
+import { EchoProvider, EchoTokens } from '@merit-systems/echo-react-sdk';
+```
+
+## Environment Variables
+- `VITE_ECHO_APP_ID` — Echo app identifier (client-side, exposed to browser)
+- Store in `.env.local` file at project root
+- Never commit `.env.local` to version control
+
+## Project Structure
+```
+src/
+  main.tsx            # App entry with EchoProvider
+  App.tsx             # Main application component
+  App.css             # Application styles
+  components/         # React components
+```
+
+## Common Patterns
+- Use Vite for dev server: `vite` or `npm run dev`
+- Build with: `tsc -b && vite build`
+- React 19 is used — leverage new features like `use()` hook
+- No server-side rendering — this is a pure client-side app
+- Use `npm dedupe` (via `postinstall`) to resolve duplicate dependency versions