Skip to main content
Glama

get_disk_usage

Read-onlyIdempotent

Lists what fills a drive or folder: a folder tree with entries by size, files by size, and file types ranked by space. Use it when a drive is full or to find what to clean up.

Instructions

Lists what fills a drive or folder, as Anabolic's Disk space page reads it: its folders as a tree (the largest entries of each level with their size and share, the rest summed into one line), the largest files under it and the file types taking the most room. Use it when a drive is full or to find what to clean up; get_disks gives the drives' size and activity, not their contents. On Windows an NTFS drive is read from its master file table when Anabolic runs as administrator (seconds for millions of files, and the whole drive even for a folder of it), otherwise folder by folder, which takes longer; progress is reported while it reads. A scan is kept for ten minutes, so asking next about a folder inside it is instant (rescan=true reads again). It only reads: nothing is deleted, the user deletes from the page.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "get_disk_usage"
  }
}

Input Schema

NameRequiredDescriptionDefault
topNoThe largest entries of each folder (1 to 50, default 8); the rest of a folder is summed into one line.
pathNoThe drive or folder ("C:\\", "D:\\Games", "/home/you"). Default: the system drive on Windows, the home folder on Linux.
depthNoFolder levels to return below it (1 to 4, default 2).
typesNoHow many file types to list, the most room first (0 to 50, default 10).
rescanNoTrue reads it again instead of using a scan of it (or of a folder above it) from the last ten minutes.
largestNoHow many of the largest files under it to list (0 to 200, default 20).

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv0.7.0

TDQS

A4.8/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Goes well beyond the readOnly/idempotent/non-destructive annotations: it discloses the Windows NTFS master-file-table fast path requiring administrator rights, the slower fallback, that even a subfolder reads the whole drive under MFT, that progress is reported, and that results are cached for ten minutes. It also restates the safety profile in user terms ("nothing is deleted, the user deletes from the page"), which is extra context on top of destructiveHint=false.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loads purpose, then usage, then behavioral caveats, so an agent can stop reading early. The long middle sentence about NTFS/MFT is dense but carries real decision-relevant information; overall length is justified, with only minor redundancy against the schema's rescan description.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a six-parameter, no-output-schema tool, the description covers output shape, defaults, performance and permission dependencies, caching, and safety. Nothing an agent needs to call it correctly or to set expectations about latency and freshness appears to be missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3, but the description adds meaning the schema lacks: the ten-minute scan reuse applies to a scan of a folder above the requested path, which explains precisely what rescan=true overrides. The remaining parameter talk (top entries summed into one line, rescan re-reads) largely mirrors the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Opens with a specific verb and resource ("Lists what fills a drive or folder") and immediately describes the return shape: a folder tree with sizes and shares, largest files, and top file types. It explicitly separates itself from get_disks ("gives the drives' size and activity, not their contents"), so an agent can pick the right sibling without opening a schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

States the triggering situations ("when a drive is full or to find what to clean up") and names the alternative tool plus the condition that rules it out. The rescan/cache sentence also tells the agent when it need not re-read, which is a usage decision rather than a parameter restatement.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.