MCP.so
ログイン

MCP Claude Code

@SDGLBL

MCP Claude Code について

MCP implementation of Claude Code capabilities and more

基本情報

カテゴリ

AI とエージェント

ライセンス

MIT

ランタイム

python

トランスポート

stdio

公開者

SDGLBL

設定

標準の設定はありません

このサーバーの README には解析可能な MCP 設定ブロックが含まれていません。インストール手順はリポジトリをご確認ください。

リポジトリ

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is MCP Claude Code?

MCP Claude Code is an MCP server that implements Claude Code capabilities, allowing Claude to directly execute instructions for modifying and improving project files. It integrates with various MCP clients including Claude Desktop.

How to use MCP Claude Code?

Install the server following the instructions in INSTALL.md, then connect it to an MCP client like Claude Desktop. The server exposes over a dozen tools (e.g., read, write, edit, run_command, dispatch_agent) for code understanding, modification, and task management.

Key features of MCP Claude Code

  • Analyze codebases with file access and pattern searching
  • Make targeted file edits with permission handling
  • Run shell commands and scripts with error handling
  • Find files and code patterns with high‑performance searching
  • Delegate complex tasks to concurrent sub‑agents
  • Support any LiteLLM‑compatible model for agent operations

Use cases of MCP Claude Code

  • Read, edit, and manage project files through conversational instructions
  • Search codebases for patterns, classes, or functions with AST context
  • Execute shell commands and scripts for build, test, or deployment tasks
  • Read and edit Jupyter notebooks, including cell outputs
  • Delegate multi‑step research or refactoring tasks to parallel sub‑agents

FAQ from MCP Claude Code

What is the difference between MCP Claude Code and Claude Code?

MCP Claude Code is an implementation of Claude Code’s capabilities using the Model Context Protocol, enabling integration with any MCP client (e.g., Claude Desktop) rather than being a standalone tool.

Does it support multiple LLM providers?

Yes. It can be configured to use any LiteLLM‑compatible model for agent operations.

How does it handle security?

The server requires permission prompts for file modifications and command execution, restricts access to specified directories, and includes input validation and error handling.

What file types and formats are supported?

It includes dedicated tools for Jupyter notebooks (read and edit with cell/output handling) and provides general file reading, writing, and searching for text‑based files.

Are there any known limitations or restrictions?

The README notes that file operations and command execution require explicit permission, and access is restricted to directories you specify for security.

コメント

「AI とエージェント」の他のコンテンツ