Skip to main content
Glama
WinDirStat MCP Server

WinDirStat MCP Server

by harezadmm

WinDirStat MCP Server 🧹📊

Model Context Protocol (MCP) Server for real-time Windows storage analysis, deep folder scanning, safety tiering, and protected cleanup operations.

Features & Exposed Tools

  1. scan_directory

    • Deep scans a folder and returns all items with size (MB/GB), file counts, and last write dates.

  2. get_largest_items

    • Ranks the top N largest files and subdirectories for any target path.

  3. categorize_safety_tiers

    • Categorizes files/folders into Tier 1 (100% Safe Cache/Temp), Tier 2 (Reviewable Downloads/Media), and Tier 3 (Protected Code/Projects).

  4. clean_safe_targets

    • Cleans temporary caches and specified paths with built-in protection guards preventing accidental deletion of project repositories.


Related MCP server: wiztree-mcp

Configuration / Installation

1. Claude Desktop Integration (claude_desktop_config.json)

Add the following entry to your %APPDATA%\Claude\claude_desktop_config.json:

{
  "mcpServers": {
    "windirstat": {
      "command": "node",
      "args": [
        "C:\\Users\\Hariz\\windirstat-mcp\\index.js"
      ]
    }
  }
}

2. Antigravity IDE / Cursor / VS Code Integration

Add to your MCP settings configuration:

{
  "mcpServers": {
    "windirstat": {
      "command": "node",
      "args": [
        "C:/Users/Hariz/windirstat-mcp/index.js"
      ]
    }
  }
}

Available Tools

4 tools
categorize_safety_tiersB

Categorize directory items into Tier 1 (100% Safe Cache/Temp), Tier 2 (Reviewable Downloads/Media), and Tier 3 (Protected Code/Projects).

Inputs: path (required).

clean_safe_targetsC

Safely delete specified temporary cache folders or files with built-in protection guards.

Inputs: targets (required), confirmAction (required).

get_largest_itemsA

Find top N largest files and subfolders under a specific path.

Inputs: path (required), limit.

scan_directoryB

Deep scan a directory and return subdirectories sorted by storage size in MB/GB with file counts.

Inputs: path (required), maxDepth, minSizeMB.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.6/5.0

    Scored across 4 tools

    Disambiguation5/5

    Each tool has a clearly distinct purpose: scanning, finding large items, categorizing by safety, and cleaning. No overlap or ambiguity.

    Naming Consistency5/5

    All tools use snake_case with a consistent verb_noun pattern (e.g., scan_directory, categorize_safety_tiers), making predictions easy.

    Tool Count5/5

    With only 4 tools, the server is well-scoped for disk cleanup. Each tool adds essential functionality without redundancy or bloat.

    Completeness4/5

    The tools cover a logical workflow from scanning to cleaning. Minor gap: no tool to view or manage individual files beyond the clean operation, but the surface is sufficient for the domain.

    Maintenance

    ActivityStale
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables natural-language queries about disk usage by driving WizTree scans and analyzing cached CSV snapshots. Users can find large files, duplicate data, folder breakdowns, and more without repeatedly rescanning the disk.
      19
      MIT
    • A
      license
      Not graded
      quality
      B
      maintenance
      Enables local AI assistants to inspect Windows disk usage with treemap and folder X-ray views, retrieve cleanup previews, and receive application removal guidance, while exposing only non-destructive commands through MCP.
      MIT
    • A
      license
      Not graded
      quality
      B
      maintenance
      Enables natural-language management of macOS disk cleanup by scanning large files, AI model weights, and caches, identifying duplicates, and safely moving files to Trash or permanently deleting them with confirmation.
      MIT