v0.2.0

This update introduces comprehensive Model Control Protocol (MCP) functionality for enhanced LLM interaction management, including prompts and resource handling systems, while also adding conversation title refresh capabilities and fixing Anthropic API compatibility issues.

Added

• Add chat title refresh functionality allowing automated regeneration of conversation titles using the LLM
• Implement Model Control Protocol (MCP) Prompts system for standardized LLM interactions and enhanced prompt management
• Add testing interface for Model Control Protocol client components to improve development reliability
• Implement resource handling system for Model Control Protocol to manage and integrate external content

Fixed

• Resolve Anthropic API compatibility issue by automatically trimming whitespace from stop sequence parameters

v0.1.1

This update introduces configuration flexibility for OpenAI API endpoints, improves system logging capabilities, and fixes a navigation issue with non-existent chat conversations, collectively enhancing both the system's stability and customization options.

Added

• Add custom endpoint configuration option for OpenAI provider, allowing connection to alternative API servers

Changed

• Enhance Model Control Protocol (MCP) client with dedicated logger for improved diagnostics and troubleshooting

Fixed

• Resolve error handling when navigating to non-existent chat conversations in the UI

v0.1.0

This release introduces a complete web-based chat interface for LLMs with support for multiple providers (Ollama, Anthropic, OpenAI, OpenRouter), persistent conversation storage, and extensive customization options. The addition of containerized deployment and structured logging improves the system's operability, while the ability to use external tools with Anthropic models extends the functional capabilities.

Added

• Add web-based user interface for chatting with Large Language Models (LLMs)
• Integrate multiple LLM providers: Ollama, Anthropic, OpenAI and OpenRouter
• Implement Bolt database for persistent storage of chat history and messages
• Add configuration file system for managing LLM provider settings
• Add customizable LLM parameters for fine-tuning model behavior
• Implement dedicated LLM instance for generating conversation titles
• Add ability to customize system prompts and conversation title generation
• Enable tools interaction capability for Anthropic models
• Display system objects (servers, tools, resources, prompts) in the user interface
• Add structured logging for improved monitoring and troubleshooting
• Include Dockerfile for containerized deployment