Skip to main content
Glama

get_weights

Medians and statistics across many ads (not single ads): the median days live, median EU reach or median versions of a format, niche, country or target, with the sample size. Use it FIRST for any question about a median, an average, a typical value or a comparison of formats: 'median days live of video ads in weight loss', 'do videos last longer than images', 'which format reaches more'. Each row: dimension (format, niche, country, target…), value (e.g. VIDEO), signal (days_live, eu_reach, versions, active, share_in_country, relaunches, test_survivor), sample size, median, top 10%, and when it was computed. Rows under 30 ads are never shown. A row may carry a 'caveat' (today on days_live, active and test_survivor: ads that died young are missing, so survival is overstated): always repeat it with the number. For formats, format_pair_weight adds weights from paired experiments (same brand, two formats launched the same week): reach ratio A vs B with a 95% interval over brands, shown from 30 brands, a winner only when the interval excludes x1; quote its note (a strong hint, not a controlled test). Costs 10 credits (0 if nothing is returned).

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_weights"
  }
}

Input Schema

NameRequiredDescriptionDefault
limitNo
nicheNo
valueNoe.g. VIDEO (for format), FR (for country)
regionNoEU or 'not measured'
signalNodays_live, eu_reach, versions, active, share_in_country, relaunches, test_survivor
countryNo
formulaNoweight formula: v5 (default, full set on the niche-page cohort) or v4 / v3 (kept until 20/11/2026)
dimensionNoformat, niche, country, target

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / limit / maximum
      Previous value: -50New value: +20
  2. Changed2 schema fields changed
    • changedInput schema / properties / formula / description
      Previous value: -"weight formula: v4 (default, on corrected niches) or v3 (kept until 20/11/2026)"New value: +"weight formula: v5 (default, full set on the niche-page cohort) or v4 / v3 (kept until 20/11/2026)"
    • changedInput schema / properties / formula / enum
      Previous value: -[
      -  "v4",
      -  "v3"
      -]New value: +[
      +  "v5",
      +  "v4",
      +  "v3"
      +]
  3. Changed1 schema field changed
    • addedInput schema / properties / formula
      Added value: +{
      +  "description": "weight formula: v4 (default, on corrected niches) or v3 (kept until 20/11/2026)",
      +  "enum": [
      +    "v4",
      +    "v3"
      +  ],
      +  "type": "string"
      +}
  4. First observed

TDQS

A4.6/5.0
Behavior4/5

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

With no annotations, the description carries the full burden and does well: it discloses the 30-ad suppression floor, the credit cost of 10 (0 on empty results), the 30-brand threshold and 'interval must exclude x1' rule for winners, and the caveat semantics. The main gaps are that it does not say what happens on partial/failed computation or whether results are cached/refreshable.

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?

It is long but front-loaded with the key scoping statement, and nearly every clause carries distinct information (suppression rules, caveats, costs, the paired-experiment path). It would read better if the caveat and format_pair_weight material were visually separated from the core row description, but there is little true filler.

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 complex 8-parameter statistical tool with no output schema and no annotations, the description supplies the missing return structure row-by-row (dimension, value, signal, sample size, median, top 10%, computed timestamp), plus the suppression and caveat rules an agent must repeat verbatim. Nothing essential to correct invocation or reporting is absent.

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 63%, and the description meaningfully compensates: it explains what `dimension` values exist (format, niche, country, target), what `signal` accepts (days_live, eu_reach, versions, active, share_in_country, relaunches, test_survivor), and gives a `value` example (VIDEO). `limit` and `region` are left unexplained beyond the schema, so it is not fully complete, but it adds real meaning over the schema for the majority of parameters.

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?

The description opens with a precise verb+resource scoping statement: 'Medians and statistics across many ads (not single ads)', which immediately distinguishes it from per-ad siblings like get_ad. It enumerates the exact signals computed (days live, EU reach, versions) and the dimensions they are broken down by, so an agent knows exactly what the tool produces without opening the 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?

Explicitly says 'Use it FIRST for any question about a median, an average, a typical value or a comparison of formats' and then supplies representative natural-language queries that map to this tool. It also names a distinct sub-path (format_pair_weight) and its specific conditions, so routing decisions are unambiguous.

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.

Resources