mcp-workspace
by MarcusJellinghaus·★ 47·Score 45
Secure MCP server providing file system, Git, and GitHub tools for AI assistants in a sandboxed project directory.
Overview
MCP Workspace Server is a Python-based MCP server that provides secure file system operations within a specified project directory. It enables AI assistants to read, write, edit, and delete files, as well as work with Git repositories and GitHub resources. The server implements strict security measures by restricting all operations to the configured project directory, preventing unauthorized file access. It includes features like structured logging, reference projects for read-only access to additional codebases, and precise file editing capabilities using exact string matching.
Try asking AI
After installing, here are 6 things you can ask your AI assistant:
When to choose this
Choose this when you need secure, sandboxed file system access for AI assistants in local development projects with support for multiple reference projects.
When NOT to choose this
Avoid if you need cloud-based file access, broader system permissions beyond project directories, or integration with services other than GitHub.
Tools this server exposes
10 tools extracted from the READMElist_directoryLists files and directories in the project directory
read_fileReads the contents of a file
save_fileCreates or overwrites files atomically
append_fileAdds content to existing files
delete_this_fileRemoves files from the filesystem
edit_fileMakes selective edits using exact string matching
move_fileMoves or renames files and directories
get_reference_projectsLists available reference projects
list_reference_directoryLists files in a reference project
read_reference_fileReads files from reference projects
Comparable tools
Installation
Installation
# Clone the repository
git clone https://github.com/MarcusJellinghaus/mcp-workspace.git
cd mcp_workspace
# Create and activate a virtual environment (optional but recommended)
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
# Install dependencies using pip with pyproject.toml
pip install -e .Running the Server
mcp-workspace --project-dir /path/to/project [--reference-project NAME=/path/to/reference]... [--log-level LEVEL] [--log-file PATH]Claude Desktop Integration
Add the following to your Claude Desktop configuration file:
{
"mcpServers": {
"filesystem": {
"command": "mcp-workspace",
"args": [
"--project-dir",
"C:\\path\\to\\your\\specific\\project",
"--reference-project",
"docs=C:\\path\\to\\documentation"
]
}
}
}FAQ
- How does the server ensure file security?
- The server restricts all file operations to the specified project directory. Attempts to access files outside this directory will result in an error. Reference projects provide read-only access only.
- Can I use multiple reference projects?
- Yes, you can specify multiple reference projects using the --reference-project argument multiple times. Each reference project is read-only and can have a unique name.
- What file operations are supported?
- The server supports list_directory, read_file, save_file, append_file, delete_this_file, edit_file, and move_file operations, plus special methods for working with reference projects.
Compare mcp-workspace with
Last updated · Auto-generated from public README + GitHub signals.