Skip to main content
Glama

Get node performance history

proxmox_get_node_stats
Read-only

Retrieve Proxmox node performance history from RRD over a chosen timeframe, summarizing min/avg/max/last per metric with a downsampled series. Inspect CPU, memory, disk, network, and load trends.

Instructions

Node performance history from the RRD database (CPU, IO wait, load, memory, swap, root disk, network, ARC...) over a timeframe, summarized as min/avg/max/last per metric plus a short downsampled series. Network values are bytes per second.

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": "proxmox_get_node_stats"
  }
}

Input Schema

NameRequiredDescriptionDefault
cfNoRRD consolidation function (default AVERAGE)
rawNoReturn the full unprocessed objects from the API (larger)
nodeNoNode name. Defaults to PROXMOX_DEFAULT_NODE, or the only node on a single-node install. See proxmox_list_nodes.
pointsNoNumber of points in the downsampled series (default 12)
metricsNoOnly these metrics, e.g. ["cpu", "memused", "netin"] (default all)
timeframeNoTime window (default hour)

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.6.1

TDQS

A3.9/5.0
Behavior4/5

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

With readOnlyHint already declaring safety, the description earns credit by disclosing the return shape (min/avg/max/last per metric plus a short downsampled series) and a normalization rule the schema does not carry: network values are bytes per second. It also flags that `raw` returns larger unprocessed payloads. It does not mention RRD retention limits (e.g., whether 'year'/'decade' have data on a fresh node) or behavior for an unknown node, which are the remaining behavioral gaps.

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?

Two dense, front-loaded sentences: what it is first, units last. No filler. The parenthetical metric list with a trailing ellipsis is slightly informal but efficiently communicates breadth.

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

Completeness4/5

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

With no output schema and a read-only annotation set, the description correctly takes on the burden of describing return values, and it does so at the right granularity for a stats endpoint. Missing only edge-case context (RRD retention across large timeframes, invalid-node handling), which is minor for a read-only query.

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

Parameters3/5

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

Schema coverage is 100%, so all six parameters are already documented, including the TAVERAGE/MAX consolidation function and the points cap. The description adds only indirect meaning — 'short downsampled series' hints at the default points value, and 'larger' contextualizes raw — but it never explains how `cf` changes the returned min/avg/max/last values, which is the one ambiguity the schema leaves open.

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?

Names a specific verb and resource ('Node performance history from the RRD database') and enumerates the exact metric families covered, plus what the payload looks like (min/avg/max/last plus a downsampled series). That framing cleanly separates it from the sibling snapshot tools such as proxmox_get_node_status and the per-guest stats tools, which report current state rather than timeframe history.

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

Usage Guidelines3/5

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

The mention of 'over a timeframe' and the enum of timeframe values implies this is for historical/trend analysis, but the description never says when to prefer it over proxmox_get_node_status or proxmox_get_vm_stats, nor when the `raw` escape hatch is warranted. Usage is inferable but not stated.

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

Deploy Server

Other Tools