OpenClient: Private AI Chat

Hi there! We've got some great new features for you in this update. ***1.6.5: • Minor bug fixes and improvements for a smoother experience. ***Latest Updates: • New widgets. • Assistant messages now render the full markdown toolkit: blockquotes, lists with nesting, task lists, horizontal rulers, code blocks with language labels, tables, and inline images all display natively in the chat. • Steer conversations mid-stream, send a follow-up message while the model is still generating; the current stream is cancelled, the partial response is preserved, and your new message continues the conversation with full context. • Connect MCP servers (Model Context Protocol) to give the model access to external tools like GitHub, databases, and more, all configured from your LiteLLM server. • Browse your MCP servers in a dedicated sheet from the chat input bar or from Settings, then drill into each server to enable or disable individual tools with a master toggle. • A new collapsible actions bar keeps the chat composer clean: attachment, web search, and MCP buttons are tucked behind a single + / × toggle, auto-collapsing after use or when sending a message. • Contextual tips introduce MCP servers when the feature becomes available, following the same unobtrusive pattern as other feature tips. • Minor bug fixes and improvements for a smoother experience. Thanks for your continued support and for helping us build the best possible LLM client together. Remember: you can suggest and vote on new features from the Feedback section in settings.