Skip to main content

A Model Context Protocol server for turning repositories into navigable MCP systems

Project description

MseeP.ai Security Assessment Badge

Dev-Kit MCP Server

PyPI - Python Version version License OS OS OS Tests Code Checks codecov Ruff Last Commit

A Model Context Protocol (MCP) server targeted for agent development tools, providing scoped authorized operations in the root project directory. This package enables secure execution of operations such as running makefile commands, moving and deleting files, with future plans to include more tools for code editing. It serves as an excellent MCP server for VS-Code copilot and other AI-assisted development tools.

dev-kit-mcp-server MCP server

Features

  • 🔒 Secure Operations: Execute operations within a scoped, authorized root directory
  • 🛠️ Makefile Command Execution: Run makefile commands securely within the project
  • 📁 File Operations: Move, Create, Rename and Delete files within the authorized directory
  • 🔄 Git Operations: Perform Git operations like status, add, commit, push, pull, and checkout
  • 🔌 MCP Integration: Turn any codebase into an MCP-compliant system
  • 🤖 AI-Assisted Development: Excellent integration with VS-Code copilot and other AI tools
  • 🔄 Extensible Framework: Easily add new tools for code editing and other operations
  • 🚀 Fast Performance: Built with FastMCP for high performance

Installation

pip install dev-kit-mcp-server

Usage

Running the Server

# Recommended method (with root directory specified)
dev-kit-mcp-server --root-dir=workdir

# With custom TOML file for predefined commands
dev-kit-mcp-server --root-dir=workdir --commands-toml=custom_commands.toml

# Alternative methods
uv run python -m dev_kit_mcp_server.mcp_server --root-dir=workdir
python -m dev_kit_mcp_server.mcp_server --root-dir=workdir

The --root-dir parameter specifies the directory where file operations will be performed. This is important for security reasons, as it restricts file operations to this directory only.

The --commands-toml parameter allows you to specify a custom TOML file for predefined commands instead of using the default pyproject.toml file. This is useful when you want to define a separate set of commands for different purposes.

Available Tools

The server provides the following tools:

File Operations

  • create_dir: Create directories within the authorized root directory
  • edit_file: Edit files by replacing lines between specified start and end lines with new text
  • move_dir: Move files and directories within the authorized root directory
  • remove_file: Delete files within the authorized root directory
  • rename_file: Rename files and directories within the authorized root directory

Git Operations

  • git_status: Get the status of the Git repository (changed files, untracked files, etc.)
  • git_add: Add files to the Git index (staging area)
  • git_commit: Commit changes to the Git repository
  • git_push: Push changes to a remote Git repository
  • git_pull: Pull changes from a remote Git repository
  • git_checkout: Checkout or create a branch in the Git repository
  • git_diff: Show diffs between commits, commit and working tree, etc.

Makefile Operations

  • exec_make_target: Run makefile commands securely within the project

Exploration Operations

  • search_files: Search for files by regex pattern in the project directory. Supports optional root directory and output length limits.
  • search_text: Search for lines in files matching a given pattern. Supports file filtering, context lines, and output length limits.
  • read_lines: Read specific lines or a range from a file. Supports line range selection and output length limits.

Predefined Commands

  • predefined_commands: Execute predefined commands from a TOML file (default: pyproject.toml under [tool.dkmcp.commands] section)

The TOML file format for predefined commands is as follows:

[tool.dkmcp.commands]
test = "uv run pytest"
lint = "ruff check"
check = "uvx pre-commit run --all-files"
doctest = "make doctest"

Each command is defined as a key-value pair where the key is the command name and the value is the command to execute. For example, when you call the predefined command "test", it will execute "uv run pytest" in the root directory.

Here's a simple example of how to define commands in a custom TOML file:

# custom_commands.toml
[tool.dkmcp.commands]
# Basic commands
hello = "echo Hello, World!"
date = "date"

# Development commands
test = "pytest"
lint = "ruff check ."
build = "python setup.py build"

Example Usage with MCP Client

from fastmcp import Client
async def example():
    async with Client() as client:
        # List available tools
        tools = await client.list_tools()

        # File Operations
        # Create a directory
        result = await client.call_tool("create_dir", {"path": "new_directory"})

        # Move a file
        result = await client.call_tool("move_dir", {"path1": "source.txt", "path2": "destination.txt"})

        # Remove a file
        result = await client.call_tool("remove_file", {"path": "file_to_remove.txt"})

        # Rename a file
        result = await client.call_tool("rename_file", {"path": "old_name.txt", "new_name": "new_name.txt"})

        # Edit a file
        result = await client.call_tool("edit_file", {
            "path": "file_to_edit.txt",
            "start_line": 2,
            "end_line": 4,
            "text": "This text will replace lines 2-4"
        })

        # Git Operations
        # Get repository status
        result = await client.call_tool("git_status")

        # Add files to the index
        result = await client.call_tool("git_add", {"paths": ["file1.txt", "file2.txt"]})

        # Commit changes
        result = await client.call_tool("git_commit", {"message": "Add new files"})

        # Pull changes from remote
        result = await client.call_tool("git_pull", {"remote": "origin", "branch": "main"})

        # Push changes to remote
        result = await client.call_tool("git_push")

        # Checkout a branch
        result = await client.call_tool("git_checkout", {"branch": "feature-branch", "create": True})

        # Makefile Operations
        # Run a makefile command
        result = await client.call_tool("exec_make_target", {"commands": ["test"]})

        # Exploration Operations
        # Search for Python files
        result = await client.call_tool("search_files", {"pattern": ".*\\.py$"})

        # Search for files in a specific directory with output limit
        result = await client.call_tool("search_files", {
            "pattern": "test.*",
            "root": "tests",
            "max_chars": 1000
        })

        # Search for text patterns in files
        result = await client.call_tool("search_text", {"pattern": "def.*test"})

        # Search in specific files with context
        result = await client.call_tool("search_text", {
            "pattern": "import",
            "files": ["main.py", "utils.py"],
            "context": 2
        })

        # Read specific lines from a file
        result = await client.call_tool("read_lines", {"file_path": "README.md", "start": 1, "end": 10})

        # Read entire file with character limit
        result = await client.call_tool("read_lines", {
            "file_path": "config.json",
            "max_chars": 500
        })

        # Predefined Commands
        # Execute a predefined command
        result = await client.call_tool("predefined_commands", {"command": "test"})

        # Execute a predefined command with a parameter
        result = await client.call_tool("predefined_commands", {"command": "test", "param": "specific_test"})

Development

Setup

# Clone the repository
git clone https://github.com/DanielAvdar/dev-kit-mcp-server.git
cd dev-kit-mcp-server

# Install development dependencies
pip install -e ".[dev]"

# Run tests
pytest

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

iflow_mcp_dev_kit_mcp_server-0.1.6.tar.gz (20.9 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

iflow_mcp_dev_kit_mcp_server-0.1.6-py3-none-any.whl (36.7 kB view details)

Uploaded Python 3

File details

Details for the file iflow_mcp_dev_kit_mcp_server-0.1.6.tar.gz.

File metadata

File hashes

Hashes for iflow_mcp_dev_kit_mcp_server-0.1.6.tar.gz
Algorithm Hash digest
SHA256 a5a2c95c9bdac439d684e371c2b2aaf2b02c8271bbe97dc54e704afe5fbfd9ab
MD5 cefb2f20082d5e5e0525cf5a6426ca35
BLAKE2b-256 49d09733ebd84c6329332914e89ddb9159c007a0d20434394fab38cfa4c6a2e5

See more details on using hashes here.

File details

Details for the file iflow_mcp_dev_kit_mcp_server-0.1.6-py3-none-any.whl.

File metadata

File hashes

Hashes for iflow_mcp_dev_kit_mcp_server-0.1.6-py3-none-any.whl
Algorithm Hash digest
SHA256 9affd857393815b712f6f51134db23f630528fffab5678a4ad98c26882658958
MD5 4c6e0d1ad074f75192a25e7260ec2b9e
BLAKE2b-256 9afa44ecdbcb00507fdb706695fafa37562296825abfe108b41aa6439e4ef5f3

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page