Provides a bridge between the Model Context Protocol (MCP) and the Language Server Protocol (LSP) to expose LSP capabilities to large language models. The MCP server starts an LSP client that connects to an LSP server, and it exposes MCP tools that translate LSP requests into MCP-friendly operations, returning results in a format that LLMs can easily understand. This enables LLMs to utilize LSP features such as hover information, code completions, and diagnostics to generate more accurate and context-aware suggestions. The repository includes a sample configuration showing how to launch the MCP server via stdio with npx, language identifiers, the path to the LSP, and optional arguments. Key tools include get_info_on_location, get_completions, get_code_actions, open_document, close_document, get_diagnostics, start_lsp, restart_lsp_server, and set_log_level for runtime control. It also offers MCP resources (lsp-diagnostics://, lsp-hover://, lsp-completions://) for real-time updates or point-in-time queries, and supports both tool-based and resource-based access. Additional features cover logging, colorized output, extensions, and a simple CLI.