$ loading_
Provides safe, directory-restricted file access for local models.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "mcp-filesystem" yet — see the docs or source repo.
Read the config file in my allowed working directory, modify it as requested, and write it back atomically without touching files outside the directory.
Returns the file contents and safely writes changes within the allowed directory.
Scan this allowed log directory, summarize the recent error logs, and save the result as a new file.
Creates a log summary file without accessing paths outside the directory.
Developers want model-assisted file edits without exposing the rest of the filesystem. This tool is suited for read/write operations inside approved directories.
In ops or automation workflows, the model can batch read, generate, or update files in approved directories while preserving path boundaries and atomic writes.
When handling external input or user-provided paths, built-in symlink protection helps reduce the risk of accidental access to sensitive files.
It provides a hardened filesystem interface that lets a local model read and write files only within user-specified directories, with symlink protection and atomic operations.
No. It can only access files within explicitly allowed directories, and paths outside those directories should not be accessed.
The description mentions symlink protection and atomic operations, which help reduce privilege-escalation and interrupted-write risks.
Securely let AI read, write, and manage local files and folders.
Securely access local files for reading, writing, and directory management.
Safely read, write, search, and manage files in a sandboxed environment.
Enable AI to read, write, search, and manage files and folders.
Securely browse and search files in a read-only directory scope.
Enhanced filesystem MCP tool for searching, reading, editing, deleting, and running commands.