Skip to main content
Glama
MCP-Claude Code Bridge

MCP-Claude Code Bridge

MCP-Claude Code Bridge Setup

1. Prerequisites

  • Docker and Docker Compose installed

  • Claude Code CLI installed (npm install -g @anthropic-ai/claude-code)

  • Anthropic API key

Related MCP server: Back-Agent MCP Server

2. Setup Steps

Create project directory:

mkdir mcp-claude-bridge
cd mcp-claude-bridge

Create the files:

Save all the provided artifacts as files in your project directory:

  • Dockerfile

  • package.json

  • server.js

  • docker-compose.yml

Set environment variable:

export ANTHROPIC_API_KEY=your_api_key_here

Build and run:

docker-compose up --build

3. Configure Claude Desktop

Add to your Claude Desktop MCP settings (~/Library/Application Support/Claude/claude_desktop_config.json on Mac):

{
  "mcpServers": {
    "claude-code-bridge": {
      "command": "node",
      "args": ["/path/to/your/mcp-claude-bridge/server.js"],
      "env": {
        "ANTHROPIC_API_KEY": "your_api_key_here"
      }
    }
  }
}

4. Test with Cat Facts App

Once connected, tell Claude Desktop:

"Create a simple cat facts app using the claude_code_task tool. The app should fetch facts from catfacts.ninja and display them in a simple web interface."

5. Available Tools

The bridge provides these tools:

  • claude_code_task: Execute coding tasks via Claude Code

  • create_project: Create new project directories

  • list_files: List files in projects

  • read_file: Read file contents

  • run_command: Execute shell commands

Example Usage

In Claude Desktop, you can now say:

  • "Create a new React app that fetches cat facts"

  • "Build a simple API server in Express"

  • "Debug this Python script for me"

The MCP server will delegate complex tasks to Claude Code while keeping everything in the Desktop interface!

Available Tools

5 tools
claude_code_taskC

Execute a coding task using Claude Code CLI

Inputs: task (required), project_path.

create_projectC

Create a new project directory

Inputs: name (required), description.

list_filesC

List files in a project directory

Inputs: project_path (required).

read_fileC

Read contents of a file

Inputs: file_path (required).

run_commandC

Run a shell command in the project directory

Inputs: command (required), project_path.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    B3.1/5.0

    Scored across 5 tools

    Disambiguation4/5

    Most tools have distinct purposes: claude_code_task for coding execution, create_project for project setup, list_files and read_file for file inspection, and run_command for shell operations. However, claude_code_task and run_command could potentially overlap if coding tasks involve shell commands, but their descriptions help differentiate them.

    Naming Consistency3/5

    The naming is mixed: claude_code_task uses a descriptive compound name, while the others follow a verb_noun pattern (create_project, list_files, read_file, run_command). This inconsistency is noticeable but still readable, as most tools use clear verbs and nouns.

    Tool Count4/5

    With 5 tools, the count is reasonable for a code bridge server, covering project creation, file operations, and command execution. It feels slightly thin for broader coding workflows, but each tool serves a clear purpose without obvious bloat.

    Completeness3/5

    The toolset covers basic project and file operations, but there are notable gaps for a code-focused domain: no tools for writing/modifying files, deleting files, or managing dependencies. This could limit agent workflows, though core inspection and execution are present.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • F
      license
      A
      quality
      Not graded
      maintenance
      Bridges Claude Desktop, Claude Code CLI, and Claude Code for VS Code, enabling developers to plan in Desktop, save specs to project directories, and invoke Claude Code CLI with seamless state sharing across interfaces.
      4
      -
    • A
      license
      Not graded
      quality
      D
      maintenance
      Connects Claude Desktop and Claude Code, enabling autonomous exchange of messages, files, and code while keeping their context windows separate.
      45 PyPI
      4
      MIT