A mobile-optimized single page chat application that uses the Anthropic Messages API with real-time streaming and MCP (Model Context Protocol) server support, featuring centralized client management and performance optimizations.
Source: https://www.val.town/x/c15r/Chat
Live server: https://c15r--7471cb243cdd11f08caa9e149126039e.web.val.run
Bundles Contextual MCP server by default: https://c15r--1564b217248f4a97a4d37bfe82cb8b19.web.val.run/mcp
- π Real-time streaming responses - See Claude's responses appear word-by-word as they're generated
- βΉοΈ Stream control - Stop streaming at any time with the stop button
- π NextSteps Auto-Execution - MCP tools can propose follow-up actions with automatic or manual execution
- π Resource Viewer - MCP tools can display file content with syntax highlighting and fullscreen support
- π§ Centralized Client Pool - Unified MCP client management for optimal performance and reliability
- π Performance Optimizations - Memoized rendering, stable references, and efficient connection reuse
- π Automatic MCP Server Status Checking - Real-time server health monitoring with detailed diagnostics
- π― Command Palette - Quick access to MCP tools and prompts via
/commands - πΎ Persistent Storage - Chat history and settings saved locally with automatic restoration
The application now uses a unified MCPClientPool that eliminates duplicate connections and provides consistent access to MCP operations:
// Single source of truth for all MCP operations
const clientPool = new MCPClientPool(connectedClients, serverConfigs);
// Unified API across all components
await clientPool.testServer(serverName);
await clientPool.fetchTools();
await clientPool.callTool(serverName, toolName, args);
- β No Duplicate Client Connections: Single client instance per server
- β Memoized Client Resolution: Cached lookups prevent unnecessary re-renders
- β Stable Reference Management: Arrays memoized to prevent reload loops
- β Optimized useEffect Dependencies: Minimal effect re-executions
- β Connection Reuse: Existing connections used for all operations
- β Multiple client instances per server (4 separate creation points)
- β Duplicate connection logic across components
- β ResourceViewer reload loops due to unstable references
- β Inconsistent error handling
- β Single client instance per server via centralized pool
- β Unified connection logic in MCPClientPool
- β Stable component rendering with memoized references
- β Consistent error handling across all components
This app implements Anthropic's official streaming guide using:
- Server-Sent Events (SSE) for real-time communication
- Proper event parsing (
message_start,content_block_delta,message_stop, etc.) - Content block reassembly for text, tool use, and thinking responses
- Error handling and keep-alive support
- Token usage tracking
- Support for all delta types (text, JSON, thinking)
βββ backend/
β βββ index.ts # Hono server with static file serving
βββ frontend/
β βββ index.html # Main HTML template
β βββ index.tsx # React app entry point
β βββ components/
β β βββ App.tsx # Main app component
β β βββ StreamingChat.tsx # Real-time streaming chat interface
β β βββ Settings.tsx # Settings flyout
β β βββ MCPStatus.tsx # MCP server status dashboard with progressive disclosure
β β βββ CommandPalette.tsx # MCP prompt command palette
β β βββ HTMLRenderer.tsx # HTML output renderer with MCP context
β β βββ NextSteps.tsx # NextSteps auto-execution component
β β βββ ResourceViewer.tsx # File/resource viewer with syntax highlighting
β β βββ Message.tsx # Individual message component
β βββ hooks/
β β βββ useAnthropicStream.tsx # Streaming chat hook
β βββ utils/
β β βββ mcpTesting.ts # Frontend MCP server testing
β β βββ mcpPrompts.ts # Frontend MCP prompt fetching
β βββ style.css # Custom styles with streaming animations
βββ shared/
β βββ types.ts # Shared TypeScript types for streaming
βββ test/ # Test suite
β βββ components/ # Component tests
β β βββ Message.test.tsx # Message component tests
β βββ utils/ # Test utilities and helpers
β β βββ test-helpers.ts # Common test utilities
β βββ setup.ts # Global test setup and configuration
β βββ README.md # Test-specific documentation
βββ package.json # NPM configuration with test scripts
βββ TESTING.md # Comprehensive testing guide
βββ README.md
The app stores configuration and chat history in localStorage:
anthropic_api_key: Your Anthropic API keyselected_model: The chosen Claude model (defaults to claude-3-5-sonnet-20241022)mcp_servers: Array of configured MCP serverschat_messages: Complete chat history (persists between page loads)auto_execute_next_steps: Global setting for NextSteps auto-execution (defaults to true)
GET /- Main application (serves frontend)GET /frontend/*- Frontend static filesGET /shared/*- Shared utility files
Note: MCP server testing and prompt fetching are now handled directly on the frontend for improved performance and reduced server load.
- Open the app at the provided URL
- Click "Settings" in the footer to configure your Anthropic API key and select your preferred Claude model
- Add/remove/toggle MCP servers as needed
- Use the "Test" button next to each MCP server to verify connectivity (shows β for success, β for errors)
- Configure NextSteps auto-execution in settings (enabled by default)
- Start chatting with Claude and watch responses stream in real-time!
- Use the stop button (βΉοΈ) to interrupt streaming at any time
- Switch models anytime in settings without losing your chat history
- Your chat history and settings will be automatically saved and restored when you return
- When MCP tools return NextSteps proposals, they'll appear as interactive pills that can auto-execute or require manual selection
The app comes pre-configured with the Val Town MCP server:
- Name: Val Town MCP
- URL:
https://c15r--81c497c63c0a11f081e29e149126039e.web.val.run/mcp - Status: Enabled by default (you'll need to add your authorization token)
- Mobile-First Design: Optimized for mobile devices with touch-friendly controls
- Real-time Streaming: Watch Claude's responses appear word-by-word as they're generated
- Stream Control: Stop streaming responses at any time with the stop button
- NextSteps Auto-Execution: MCP tools can propose follow-up actions that execute automatically after a countdown or require manual selection. See NextSteps-README.md for implementation details.
- Resource Viewer: MCP tools can display file and resource content with syntax highlighting, fullscreen support, and intelligent rendering based on file type
- MCP Integration: Support for multiple Model Context Protocol servers
- Enhanced MCP Display: Collapsible sections for tool calls and results with proper JSON formatting
- Unobtrusive Footer Controls: Settings (βοΈ) and clear chat (ποΈ) buttons positioned in the footer alongside the input
- Persistent Chat History: All messages automatically saved to localStorage and restored on page load
- Persistent Settings: All configuration stored in browser localStorage
- Auto-scroll: Messages automatically scroll to bottom during streaming
- Auto-resize: Input field grows with content
- Error Handling: Clear error messages for API issues with stream recovery
- Loading States: Visual feedback during API calls and streaming
- Structured Responses: MCP tool use and results are displayed in organized, collapsible sections
- Clean Interface: Maximized chat area with no header, footer contains all controls
- Streaming Animations: Visual cursor and highlighting for active streaming responses