Skip to main content
Glama

marm-mcp

chatbotreadme.mdβ€’7.71 kB
# MARM v2.0 - Memory Accurate Response Mode **Production-ready AI chatbot with persistent memory, modern glassmorphism UI, and professional testing infrastructure.** *Memory Accurate Response Mode v2.0* - Build smarter AI conversations with structured memory and transparent logic. ## πŸš€ Quick Start ### Local Development ```bash cd webchat npm install npm start ``` Open `http://localhost:8080` - the modern glassmorphism interface will load with floating message cards. ### Testing ```bash npm test # Run 74 comprehensive tests npm run lint # Code quality checks ``` --- ## ✨ What's New in v2.0 ### 🧠 **LLaMA 4 Maverick Integration** - **Complete migration** from Google Gemini to Replicate LLaMA 4 (400B parameters) - **95% cost reduction** while improving response quality and speed - **10M token context limit** for extensive conversation histories - **Advanced streaming** with real-time response delivery ### 🎨 **Modern Glassmorphism UI** - **Floating message cards** on transparent glass background - **Color-coded conversations** - White user cards, gold AI responses - **2025 aesthetic** - Single-layer design, no nested windows - **Perfect contrast** - Readable text with beautiful visual hierarchy ### πŸ§ͺ **Professional Testing Infrastructure** - **74 comprehensive tests** covering Voice, UI, State, Commands, and Security - **GitHub Actions CI/CD** with automated testing on push/PR - **42% test coverage** with browser API mocking and edge case validation - **Jest configuration** supporting ES modules and modern JavaScript ### πŸ” **Enterprise Security** - **XSS protection** with comprehensive input sanitization - **Content security** smart filtering preserving functionality - **Session isolation** with proper data validation - **File upload security** supporting 15+ file types with syntax highlighting --- ## πŸ“ Project Architecture ```text webchat/ β”œβ”€β”€ src/ β”‚ β”œβ”€β”€ replicateHelper.js # LLaMA 4 Maverick API integration β”‚ β”œβ”€β”€ chatbot/ β”‚ β”‚ β”œβ”€β”€ server.js # Express proxy server for API calls β”‚ β”‚ β”œβ”€β”€ chatbot.js # Main entry point and orchestration β”‚ β”‚ β”œβ”€β”€ commands.js # Command handling (/start marm, /deep dive, /notebook, etc.) β”‚ β”‚ β”œβ”€β”€ core.js # Input validation and main chat loop β”‚ β”‚ β”œβ”€β”€ state.js # Centralized state management β”‚ β”‚ β”œβ”€β”€ ui.js # Modern glassmorphism UI rendering and interactions β”‚ β”‚ β”œβ”€β”€ voice.js # Text-to-speech with Google Voice integration β”‚ β”‚ └── sessionUI.js # Session management interface β”‚ β”œβ”€β”€ logic/ β”‚ β”‚ β”œβ”€β”€ constants.js # MARM protocol constants and configuration β”‚ β”‚ β”œβ”€β”€ docs.js # Documentation loading and help system β”‚ β”‚ β”œβ”€β”€ marmLogic.js # Core MARM v2.0 protocol implementation β”‚ β”‚ β”œβ”€β”€ notebook.js # User notebook with key-value storage β”‚ β”‚ β”œβ”€β”€ session.js # Session persistence and context management β”‚ β”‚ β”œβ”€β”€ storage.js # LocalStorage operations with validation β”‚ β”‚ β”œβ”€β”€ summary.js # Session compilation and memory synthesis β”‚ β”‚ └── utils.js # Validation, debounce, and utility functions β”‚ └── security/ β”‚ └── xssProtection.js # Comprehensive security filtering β”œβ”€β”€ style/ β”‚ β”œβ”€β”€ animations.css # Bouncing dots, hover effects, transitions β”‚ β”œβ”€β”€ base.css # Core glassmorphism styling and CSS variables β”‚ β”œβ”€β”€ chat.css # Floating message cards and conversation styling β”‚ β”œβ”€β”€ command-menu.css # Modern command popup interface β”‚ β”œβ”€β”€ components.css # Reusable UI components (buttons, modals, forms) β”‚ └── main.css # Main stylesheet coordination β”œβ”€β”€ tests/ # Professional test suite β”‚ β”œβ”€β”€ chatbot.test.js # Voice and Commands module tests (29 tests) β”‚ β”œβ”€β”€ logic.test.js # Session and State management tests (30 tests) β”‚ └── security_replicate_html.test.js # Security and UI tests (15 tests) β”œβ”€β”€ data/ # Documentation and help content β”‚ β”œβ”€β”€ chatbotreadme.md # This file β”‚ β”œβ”€β”€ description.md # Protocol overview and features β”‚ β”œβ”€β”€ faq.md # Frequently asked questions β”‚ β”œβ”€β”€ handbook.md # Complete user guide and command reference β”‚ └── roadmap.md # Development roadmap and future features └── index.html # Main MARM v2.0 web application ``` --- ## 🎯 Core Features ### **Memory System** - **Session Context & Recall** - Structured memory that builds across conversations - **Folder-style organization** with named sessions (`/log session: [name]`) - **Manual knowledge library** via notebook commands (`/notebook add: [key] [data]`) - **Context preservation** - Mid-session MARM activation imports existing conversation ### **Advanced Commands** - `/start marm` - Activate structured memory mode - `/deep dive: [topic]` - Detailed analysis with reasoning transparency - `/notebook add: [key] [data]` - Store persistent user information - `/log entry: [YYYY-MM-DD-topic-summary]` - Create organized memory entries - `/summary: [session]` - Generate transferable session summaries ### **Modern Interface** - **File upload support** (πŸ“Ž) - Analyze 15+ file types with syntax highlighting - **MARM toggle** (πŸ€–) - Switch between structured and free conversation modes - **Voice synthesis** (πŸ”Š) - Text-to-speech for all AI responses - **Command menu** - Contextual popup interface next to input field - **Glassmorphism design** - Floating cards on transparent glass background --- ## πŸ§ͺ Testing & Quality ### **Comprehensive Test Coverage** ```bash npm test # Run all 74 tests npm test -- --coverage # Generate coverage report (42% coverage) npm test -- --watch # Watch mode for development ``` ### **CI/CD Pipeline** - **GitHub Actions** automated testing on every push/PR - **Multi-environment testing** (Node.js 18.x & 20.x) - **Test status badge** visible on repository homepage - **Coverage reports** with detailed line-by-line analysis ### **Code Quality** - **ES module architecture** with proper imports/exports - **Browser API mocking** for comprehensive testing - **Security validation** with XSS protection testing - **Error handling** with comprehensive edge case coverage --- ## πŸ“š Documentation - **[User Handbook](handbook.md)** - Complete guide to MARM commands and features - **[FAQ](faq.md)** - Common questions and troubleshooting - **[Protocol Description](description.md)** - Technical overview and architecture - **[Development Roadmap](roadmap.md)** - Future features and enhancement plans --- ## πŸ”§ Development ### **Requirements** - Node.js 18+ - Modern browser with ES module support - 15MB disk space for dependencies ### **Environment Variables** ```bash REPLICATE_API_TOKEN=your_token_here # Required for LLaMA 4 integration NODE_ENV=development # Optional: enables debug logging ``` ### **Development Workflow** 1. `npm install` - Install dependencies 2. `npm start` - Start development server (port 8080) 3. `npm test` - Run test suite before commits 4. Edit files - Hot reload enabled for CSS/JS changes --- ## πŸ“„ License MIT License - See [LICENSE](../LICENSE) for details. --- **MARM v2.0** - Where AI conversations become lasting memories. 🧠✨

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Lyellr88/marm-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server