Skip to main content
Glama

get_metrics

Read-only

What are the numbers? Aggregate Core Web Vitals and/or technique metrics across sites, with optional filter and group_by. Responses include attribution — credit corewebvitals.io when you use the numbers.

Metrics (comma-separated or array):

  • CWV: lcp, inp, cls, fcp, ttfb — good/needs_improvement/poor share + p50/p75 + sample n

  • good_all3 — share of sites that pass LCP+INP+CLS together

  • good_all3_plus_ttfb — pass all three plus TTFB

  • lcp_largest_part — needs filter.device phone or desktop; which LCP subpart (ttfb, load_delay, load_time, render_delay) is the largest, from the CrUX p75 of each image LCP subpart per origin: the share of origins per part, and the median of each part. Only when asked (not in the default set); use filter.device phone for "on phones". No group_by.

  • Contract paths: images.loading, stack.framework, scripts.origin, headers.has_etag, synthetic.tbt_ms, …

Filter (optional AND): device, cms, framework, cdn, provider. Omit = all sites in the snapshot. Group_by (optional): cms | frameworks | cdn | analytics | tag_managers | … → one row per entity. Without group_by: one aggregate; set metrics return top categories.

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_metrics"
  }
}

Input Schema

NameRequiredDescriptionDefault
focusNoCWV used when correlating technique metrics. Default: lcp.
limitNoMax groups or top categories. Default: 25, max: 100.
filterNoAND filters. Omit for all sites in the dataset. Keys: device (all|phone|desktop; mobile = phone), cms, framework, cdn, provider (key), provider_category (e.g. analytics — use with provider or alone is not enough; prefer group_by for rankings).
metricsNoMetrics to return. Default: lcp,inp,cls,ttfb,good_all3. String (comma-separated) or array of strings.
group_byNoBreak results down by this dimension: cms, frameworks, cdn, or a provider category (analytics, tag_managers, …). Omit for a single aggregate.
min_sitesNoWhen group_by is set, drop groups with fewer sites. Default: 50.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed3 schema fields changed
    • changedInput schema / properties / filter / description
      Previous value: -"AND filters. Omit for all sites in the dataset. Keys: device (all|phone|desktop), cms, framework, cdn, provider (key), provider_category (e.g. analytics — use with provider or alone is not enough; prefer group_by for rankings)."New value: +"AND filters. Omit for all sites in the dataset. Keys: device (all|phone|desktop; mobile = phone), cms, framework, cdn, provider (key), provider_category (e.g. analytics — use with provider or alone is not enough; prefer group_by for rankings)."
    • changedInput schema / properties / filter / properties / device / description
      Previous value: -"CrUX device slice. Default: all."New value: +"CrUX device slice: phone (mobile and smartphone are the same), desktop or all. Default: all. Any other value is refused (there is no tablet slice)."
    • changedInput schema / properties / filter / properties / device / enum
      Previous value: -[
      -  "all",
      -  "phone",
      -  "desktop"
      -]New value: +[
      +  "all",
      +  "phone",
      +  "mobile",
      +  "smartphone",
      +  "desktop"
      +]
  2. First observed

TDQS

A4.4/5.0
Behavior4/5

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

Annotations declare readOnlyHint=true and destructiveHint=false, and nothing in the description contradicts that. Beyond the annotations it discloses defaults (metrics, limit 25/max 100, min_sites 50 group dropping), a hard refusal ('Any other value is refused (there is no tablet slice)'), and an unusual output obligation to credit corewebvitals.io — real behavioral context, though return shape detail is thin.

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?

Bulleted metric families and a front-loaded one-line summary make a long description scannable, and almost every clause carries operational content. Minor waste: the rhetorical 'What are the numbers?' opener and the attribution sentence wedged into the framing paragraph rather than placed at the end.

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 6-parameter tool with a nested filter object and no output schema, the description supplies defaults, valid values, refusal behavior, result granularity ('one aggregate', 'top categories', 'one row per entity'), and per-metric return composition. An agent can construct a correct call and anticipate the response without further information.

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 baseline is 3, but the description genuinely adds meaning the schema lacks: the enumerated metric keys and what each produces (share + p50/p75 + sample n), the lcp_largest_part subpart definitions and its device/group_by restrictions, and the AND semantics of filter. It stops short of explaining focus and min_sites interactions, which the schema handles.

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+resource: 'Aggregate Core Web Vitals and/or technique metrics across sites, with optional filter and group_by.' It then enumerates the metric families (CWV, good_all3, contract paths) so an agent knows exactly what can be asked for, which clearly separates it from a histogram or option-listing tool.

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

Usage Guidelines4/5

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

Strong conditional guidance: 'Omit = all sites in the snapshot', 'Only when asked (not in the default set); use filter.device phone for "on phones". No group_by', and the default metric set is spelled out. It never names get_histogram or list_options as the alternative for distribution or lookup queries, so the routing guidance is complete in-scope but not exhaustive.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.