Allow AI assistants to read, write, and manage files on your local machine within controlled directories.
npx -y @modelcontextprotocol/server-filesystem /path/to/allowed/dir
Paste this into your MCP client configuration (Cursor, Windsurf, etc).
Grant your AI assistant controlled access to your local filesystem.
npx -y @modelcontextprotocol/server-filesystem /Users/you/projects
{
"mcpServers": {
"filesystem": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-filesystem", "/path/to/dir"]
}
}
}
MCP (Model Context Protocol) servers extend your AI assistant with external tools and data sources. Follow these steps to install Filesystem MCP Server in your preferred AI coding tool.
claude_desktop_config.jsonFull GitHub repository access for reading code, managing issues, reviewing PRs, and repository operations.
Query and manage SQLite databases for local data storage and analysis.
Full GitHub repository access. Read code, manage issues, and review PRs.