MCP Web UI
by MegaGrindStone
MCP Web UI is a web-based user interface that serves as a Host within the Model Context Protocol (MCP) architecture. It provides a user-friendly interface for interacting with Large Language Models (LLMs) while managing context aggregation and coordination.
Last updated: N/A
What is MCP Web UI?
MCP Web UI is a web-based interface designed to simplify and enhance interactions with AI language models by providing a unified platform for multiple LLM providers and robust context handling using the MCP protocol.
How to use MCP Web UI?
To use MCP Web UI, clone the repository, configure your environment by setting API keys and modifying the config.yaml file, and then run the application either locally or using Docker. The configuration file allows customization of server settings, prompt configurations, and LLM provider-specific parameters.
Key features of MCP Web UI
Multi-Provider LLM Integration (Anthropic, OpenAI, Ollama, OpenRouter)
Intuitive Chat Interface
Real-time Response Streaming via Server-Sent Events (SSE)
Dynamic Configuration Management
Advanced Context Aggregation
Persistent Chat History using BoltDB
Flexible Model Selection
Use cases of MCP Web UI
Interacting with multiple LLMs through a single interface
Managing and aggregating context for improved AI responses
Developing AI-powered chat applications with real-time streaming
Experimenting with different LLM configurations and parameters
FAQ from MCP Web UI
What LLM providers are supported?
What LLM providers are supported?
The UI supports Anthropic (Claude models), OpenAI (GPT models), Ollama (local models), and OpenRouter (multiple providers).
How do I configure API keys?
How do I configure API keys?
API keys can be set as environment variables (ANTHROPIC_API_KEY, OPENAI_API_KEY, OPENROUTER_API_KEY) or within the config.yaml file.
Can I customize the system prompt?
Can I customize the system prompt?
Yes, the systemPrompt in the config.yaml file allows you to define the default system prompt for the AI assistant.
How does context aggregation work?
How does context aggregation work?
The UI uses the Model Context Protocol (MCP) to manage and aggregate context between clients and servers, enabling more coherent and relevant AI responses.
Where is the chat history stored?
Where is the chat history stored?
Persistent chat history is stored using BoltDB.