- Home
- MCP servers
- Tafa
Tafa
- javascript
1
GitHub Stars
javascript
Language
6 months ago
First Indexed
2 months ago
Catalog Refreshed
Documentation & install
Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.
Installation
Add the following to your MCP client configuration file.
Configuration
View docs{
"mcpServers": {
"n3urax-tafa-mcp": {
"command": "tafa-mcp",
"args": [
"/Users/username/Documents",
"/Users/username/Desktop",
"/Users/username/Downloads"
]
}
}
}Tafa MCP Server is a production-ready server that exposes comprehensive file system management operations for seamless integration with Claude Desktop. It provides safe, sandboxed file and directory actions, robust backups, and powerful search capabilities, enabling you to perform complex file tasks from Claude with confidence.
How to use
You will run the MCP server locally or in a container and connect Claude Desktop to it. Start by launching the MCP server with the directories you want to manage. Once it is running, configure Claude Desktop to point at the MCP server so you can issue natural-language file commands that translate into filesystem operations.
Typical capabilities you can leverage include reading and writing files, creating directories, listing and deleting contents, and performing content searches. You can ask Claude to read a file, create notes, search for JavaScript files in a project, or locate duplicate files by content hash. The server enforces security boundaries to ensure operations stay within your allowed directories and automatically backups destructive actions.
How to install
Prerequisites: you need a supported runtime to run the MCP server and a client that can connect to it. A typical setup uses Node.js and npm.
# Quick install (global CLI)
npm install -g tafa-mcp
# Manual install (clone, install, and prepare)
git clone https://github.com/your-org/tafa-mcp.git
cd tafa-mcp
npm install
npm run install-global
Additional configuration and security notes
Configure Claude Desktop to connect to the MCP server and specify the directories you want to manage. The following examples show how to configure a local runtime and a Docker-based runtime.
## Claude Desktop configuration (local runtime)
{
"mcpServers": {
"tafa-mcp": {
"command": "tafa-mcp",
"args": [
"/Users/username/Documents",
"/Users/username/Desktop",
"/Users/username/Downloads"
]
}
}
}
## Claude Desktop configuration (Docker runtime)
{
"mcpServers": {
"tafa-mcp": {
"command": "docker",
"args": [
"run", "-i", "--rm",
"--mount", "type=bind,src=/Users/username/Documents,dst=/workspace/Documents",
"--mount", "type=bind,src=/Users/username/Desktop,dst=/workspace/Desktop",
"tafa-mcp",
"/workspace"
]
}
}
}
Starting the server
Run the server directly from the command line so Claude can connect to it. You can start the server and point Claude at the directories you want to manage.
tafa-mcp /path/to/directory1 /path/to/directory2
Notes on usage with Claude
After you configure Claude Desktop, you can use natural language prompts to perform file operations. Examples include asking to read a file, create a new file with sample content, search for files by name, and find duplicates by content hash.
Available tools
read_file
Read complete file contents using UTF-8 encoding.
write_file
Create a new file or overwrite an existing one with an automatic backup.
append_file
Append content to an existing file.
copy_file
Copy a file to a new location.
move_file
Move or rename a file.
delete_file
Delete a file with optional backup and confirmation.
get_file_info
Retrieve detailed file metadata including size, dates, permissions, and hash.
create_directory
Create a directory, including any missing parent directories.
list_directory
List contents of a directory with indicators for files and subdirectories; supports recursion.
delete_directory
Delete a directory with backup and confirmation.
get_directory_info
Get metadata and statistics for a directory.
search_files
Recursively search for files by name pattern within a directory.
search_content
Search within file contents (grep-like functionality) across files.
find_duplicates
Find duplicate files by comparing content hashes.
get_server_info
Provide server information and available tools.