codeTree

Overview

codeTree is an MCP server that provides AI coding agents with structured code understanding through tree-sitter parsing. It offers 23 tools for code analysis including skeleton extraction, symbol resolution, call graph generation, complexity analysis, dead code detection, and change impact analysis. The server supports 10 popular programming languages including Python, JavaScript, TypeScript, Go, Rust, Java, C, C++, and Ruby, with a one-command installation process via uv.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• get_file_skeletonget_file_skeleton(file_path)

  Get classes, functions, methods with line numbers + doc comments

  Try:Show me the structure of this fileTry:List all functions in utils.py
• get_symbolget_symbol(file_path, symbol_name)

  Get full source of a function or class

  Try:Show me the complete definition of Calculator.addTry:Get the source code for authenticate_user
• find_referencesfind_references(symbol_name)

  Find all usages of a symbol across the repository

  Try:Where is calculate_total used?Try:Show all references to DatabaseConnection
• get_call_graphget_call_graph(file_path, function_name)

  Get what a function calls and what calls it

  Try:Show the call graph for main()Try:What does payment_processor call?
• get_complexityget_complexity(file_path, function_name)

  Get cyclomatic complexity breakdown

  Try:How complex is the validate function?Try:Show complexity metrics for render_component
• search_symbolssearch_symbols(query?, type?, parent?)

  Flexible symbol search with filters

  Try:Find all functions that handle authenticationTry:Search for classes in auth.py
• get_repository_mapget_repository_map(max_items?)

  Get compact repo overview: languages, entry points, hotspots

  Try:Give me an overview of this codebaseTry:What are the main entry points?
• get_change_impactget_change_impact(symbol_query?, diff_scope?)

  Analyze impact of code changes with risk levels

  Try:What will break if I change User class?Try:Show impact of modifying the payment API
• analyze_dataflowanalyze_dataflow(file_path, function_name, mode?)

  Analyze variable dataflow or taint tracing

  Try:Trace how user input flows in authTry:Show dataflow in payment processing
• get_dependency_graphget_dependency_graph(file_path?, format?)

  Get file-level dependency graph as Mermaid or list

  Try:Show dependencies for main.pyTry:Visualize module relationships
• find_testsfind_tests(file_path, symbol_name)

  Find test functions for a specific symbol

  Try:Find tests for the Calculator classTry:Show tests for validate_email
• suggest_docssuggest_docs(file_path?, symbol_name?)

  Find undocumented functions with context for doc generation

  Try:Find functions that need documentationTry:Suggest docs for payment module

Comparable tools

Installation

claude mcp add codetree -- uvx --from mcp-server-codetree codetree --root .

{
  "mcpServers": {
    "codetree": {
      "command": "uvx",
      "args": ["--from", "mcp-server-codetree", "codetree", "--root", "/path/to/your/project"]
    }
  }
}

FAQ

What makes codeTree different from reading files directly?

codeTree reduces token consumption by up to 25x by providing structured code understanding without reading entire files. Agents can ask for specific symbols, skeletons, or relationships rather than processing raw code.

Does codeTree replace LSP servers?

No, codeTree complements LSP servers by providing a lightweight, language-agnostic approach to code understanding through MCP. It's designed specifically for AI agents rather than human developers.

Compare codeTree with