filesystem-mcp
Links
README
From the repo.
MCP FileSystem Tools
An MCP (Model Context Protocol) server that provides tools for AI assistants to interact with the file system through a single, efficient tool call.
Primary Purpose
This project is designed with one core goal: reducing multiple tool calls to a single, efficient operation when interacting with the file system.
Traditional approaches often require multiple separate calls to explore directories or read files, which is inefficient and increases latency. This project solves this problem by providing:
-
list_structure: Retrieves an entire directory structure in one call, eliminating the need for repeated directory listing operations. -
read_files: Reads multiple files in a single operation, replacing numerous individual file read requests.
Both tools require absolute paths and follow strict usage guidelines in rules.md to ensure optimal performance.
Key Benefits
- Maximum Efficiency: Complete complex file operations with just one tool call
- Reduced Latency: Eliminate the overhead of multiple back-and-forth communications
- Simplified Workflows: Streamline AI interactions with the file system
- Consistent Results: Standardized approach to path handling and error management
Recommended Agents
Based on testing, this project works optimally with advanced AI agents such as Gemini 2.5 Pro and Claude 3.7. These agents are adept at leveraging the efficient, single-call nature of the provided tools.
License
MIT
Collected info
- ★ 22 stars
- ⎇ 4 forks
- Language: JavaScript
- Source updated: 8/7/2026
Config for your environment
Replace {MCP_ENDPOINT_URL} with this MCP’s endpoint URL (from its repo or docs above). No API key — you connect directly.
Tool
OS
Config file: ~/.cursor/mcp.json
{
"mcpServers": {
"mcp-server": {
"url": "{MCP_ENDPOINT_URL}"
}
}
}Paste into mcpServers in the config file. Restart Cursor after saving.
If this MCP is also published on mcpchannel.ai, you can subscribe from Browse and use the gateway config there instead.