Code Index MCP enables large language models to index, search, and analyze code in project directories. It provides tools for navigating file structures, searching for patterns in code, generating file summaries, and analyzing code complexity across multiple programming languages. The implementation uses UV for dependency management, stores project settings persistently in a .code_indexer folder, and integrates easily with Claude Desktop through a simple configuration process, making it ideal for developers who need to analyze codebases without leaving their AI assistant interface.
No reviews yet. Be the first to review!
Sign in to join the conversation
Highlight AIInitialize indexing for a project directory.
Rebuild the shallow file index after file changes.
Generate the full symbol index used by deep analysis.
View current project configuration and status.
Smart search with regex, fuzzy matching, file filtering, and paginated results (10 per page by default).
Locate files using glob patterns (e.g., **/*.py).
Analyze file structure, functions, imports, and complexity (requires deep index).
Check file watcher status and configuration.
Enable/disable auto-refresh and configure settings.
Set up storage directory for index data.
Verify index storage location and permissions.
Reset all cached data and configurations.
Re-detect available search tools (ugrep, ripgrep, etc.).
