mcp-server-filesystem

MCP serverFiles & storage

mcp-server-filesystem is a filesystem MCP server that gives an AI agent access to local files. Once connected, the agent can read, write, and manage files on the machine it runs on, so tasks involving documents, code, or data can be handled directly instead of being pasted in by hand.

Unavailable. This server has no hosted endpoint yet, so ahel can't serve it.

Have an AI agent setup that supports MCP servers.

What your AI can do with it

  • Read files from the local filesystem
  • Write and update files on the local machine
  • Manage files and directories through MCP tools
  • Expose filesystem access to any MCP-compatible agent

Getting started

  1. Have an AI agent setup that supports MCP servers.
  2. Add mcp-server-filesystem to the agent's MCP server configuration.
  3. Grant the server access to the directories you want the agent to work with.
  4. Restart or reload the agent so the filesystem tools become available.

From the project's README

As published by bytedance/ui-tars-desktop in README.md. This repository publishes several servers, so only the opening is shown here.

Introduction

English | 简体中文

TARS* is a Multimodal AI Agent stack, currently shipping two projects: Agent TARS and UI-TARS-desktop:

Shortened here. Read the whole README on GitHub.

Signals

GitHub stars
39k
Forks
4k
Last commit
Sep 2026
Weekly downloads
382

Questions

What is a filesystem MCP server?
It is an MCP server that exposes filesystem operations to an AI agent, letting the agent read, write, and manage local files through standard MCP tools.
Is it safe to give an agent access to my files?
Access is limited to the directories you configure when setting up the server, so you control which parts of the filesystem the agent can reach.
Do I need anything special to run it?
You need an agent environment that supports MCP servers, plus configuration pointing the server at the directories you want to share.
Advanced
Delivery
mcp-server-filesystem MCP server → your ahel connector (mcp.ahel.ai) → your AI.
Item type
mcp-server
Key
io-github-bytedance-mcp-server-filesystem
Source
github.com/bytedance/ui-tars-desktop