About FileMoverMcp
FileMoverMcp stages file and directory moves for review before execution, for CLI users and AI agents. Its session workflow supports previews and shared state between CLI and MCP modes.
Key features
- 01Stage file and directory moves
- 02Preview pending moves
- 03Commit or cancel operations
- 04Create destination directories automatically
- 05Restrict paths to the session base
- 06Require explicit overwrite permission
Use cases
- 01Review a batch of file moves before execution
- 02Move files while preserving directory structure during refactoring
- 03Stage moves in the CLI and commit them through MCP
MCP support
When run as an MCP server, it exposes tools for initializing sessions, staging moves, previewing, committing, and canceling. The CLI and MCP modes share session storage.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEFileMoverMcp
A file mover tool that stages and executes file moves safely. Works as both a CLI tool and an MCP (Model Context Protocol) server for AI agents.
Installation
dotnet tool install --global FileMoverMcp
dotnet tool update --global FileMoverMcp
dotnet tool uninstall --global FileMoverMcp
After installation, the fm command will be available globally.
MCP setup
To register as an MCP tool in Claude Code:
claude mcp add filemover -- fm --mcp
For Cursor or other MCP clients, add this to your MCP configuration:
{
"mcpServers": {
"filemover": {
"command": "fm",
"args": ["--mcp"]
}
}
}
Usage
The workflow is: initialize a session, stage moves, preview, then commit (or cancel).
fm init [path] # Initialize at path (default: current directory)
fm mv <source> <dest> # Stage a file or directory move
fm mv <source> <dest> --overwrite # Allow overwriting existing destination
fm preview # See all staged moves
fm commit # Execute all staged moves
fm cancel # Discard all staged moves
fm help # Show help information
Examples
# Move files on desktop
fm init "c:/users/adam/desktop"
fm mv "old-folder/file.png" "new-folder/file.png"
fm mv "document.txt" "archive/document-backup.txt"
fm preview
fm commit
# Move an entire directory
fm init
fm mv "source-folder" "dest-folder"
fm commit
# Overwrite an existing file
fm init
fm mv "source.txt" "destination.txt" --overwrite
fm commit
Behavior
Directories are auto-created
If the destination path includes directories that don't exist yet, they will be created automatically. For example, fm mv "file.txt" "a/b/c/file.txt" will create a/b/c/ if needed.
Files and directories are auto-detected
When you run fm mv, the tool checks whether the source is a file or a directory and handles it accordingly. Directory moves preserve the full internal structure, including nested subdirectories.
Overwrite is opt-in
If the destination already exists (file or directory), the move will be rejected unless you pass --overwrite. With --overwrite, the existing destination is replaced entirely.
Paths are sandboxed
All paths are relative to the session's base directory. Attempting to reference paths outside it will be rejected.
Sessions are shared
CLI and MCP mode share the same session storage, so you can initialize a session in one mode and commit in the other.
As MCP Server
fm --mcp
When running as an MCP server, the following tools are available:
fm_init(path?: string)- Initialize a sessionfm_move(source: string, destination: string, overwrite: boolean)- Stage a movefm_preview()- Preview staged movesfm_commit()- Execute staged movesfm_cancel()- Cancel sessionfm_help()- Get help
Development
git clone <repository-url>
cd FileMoverMcp
dotnet build FileMoverMcp.sln
dotnet test FileMoverMcp.sln
To run as MCP server during development:
dotnet run --project FileMoverMcp.Cli/FileMoverMcp.Cli.csproj -- --mcp
To package:
dotnet pack FileMoverMcp.Cli/FileMoverMcp.Cli.csproj --configuration Release
License
MIT License
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.