MCP.so
Sign In

Claude Code MCP Server

@ebeloded

Repository

About Claude Code MCP Server

An MCP server that allows AI tools to interact with Claude Code programmatically with session continuity and async execution support.

Basic information

Category

AI & Agents

License

MIT

Runtime

node

Transports

stdio

Publisher

ebeloded

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "claude": {
      "command": "npx",
      "args": [
        "claude-mcp"
      ]
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is Claude Code MCP Server?

An MCP server that bridges AI tools with Claude Code, enabling agent-in-agent workflows with conversation continuity and both synchronous and asynchronous execution. It lets AI systems programmatically start or resume Claude Code conversations, monitor long-running tasks, and branch conversations from any previous response.

How to use Claude Code MCP Server?

Install via npx or local clone, then add the MCP server configuration to your client’s MCP config file (e.g., ~/.cursor/mcp.json). Use the tools start, resume, status, and cancel to interact with Claude Code. Requires Node.js 18+ and the Claude Code CLI installed and configured.

Key features of Claude Code MCP Server

  • Fresh conversations via start tool (sync/async)
  • Resume conversations via resume tool (sync/async)
  • Working directory support for different projects
  • Async task monitoring with elapsed time and cancellation
  • Conversation branching from any previous Response ID
  • Automatic Claude CLI detection and permission bypass

Use cases of Claude Code MCP Server

  • Starting new code analysis tasks in different project directories
  • Continuing a conversation with Claude to build on prior analysis
  • Running long‑lived code reviews asynchronously while other work proceeds
  • Branching a conversation from an earlier response to explore alternative solutions

FAQ from Claude Code MCP Server

What tools does the MCP server expose?

It exposes start, resume, status, and cancel tools for managing Claude Code conversations and tasks.

How does asynchronous execution work?

The start and resume tools accept an async parameter (defaults to true) to run non‑blocking; the server returns a task ID for status monitoring and cancellation.

What are the runtime dependencies?

Node.js 18+ and the Claude Code CLI installed and run once with --dangerously-skip-permissions to accept terms.

Does it always bypass permissions?

Yes, it uses --dangerously-skip-permissions for full functionality. Future versions will support configurable permission modes.

How can I customize behavior with environment variables?

Set CLAUDE_CLI_NAME (default “claude”), MCP_CLAUDE_DEBUG for verbose logging, or MCP_NOTIFICATIONS to disable macOS notifications.

Comments

More AI & Agents MCP servers

Just Prompt - A lightweight MCP server for LLM providers

disler

just-prompt is an MCP server that provides a unified interface to top LLM providers (OpenAI, Anthropic, Google Gemini, Groq, DeepSeek, and Ollama)

733

mcp-hfspace MCP Server 🤗

evalstate

MCP Server to Use HuggingFace spaces, easy configuration and Claude Desktop mode.

387

Solon Ai

opensolon

Java AI application development framework (supports LLM-tool,skill; RAG; MCP; Agent-ReAct,Team-Agent). Compatible with java8 ~ java25. It can also be embedded in SpringBoot, jFinal, Vert.x, Quarkus, and other frameworks.

416

MCP-LLM Bridge

patruff

Bridge between Ollama and MCP servers, enabling local LLMs to use Model Context Protocol tools

976

meGPT - upload an author's content into an LLM

adrianco

Code to process many kinds of content by an author into an MCP server

287