{
  "id": "N01-HUM",
  "url": "https://cartforagents.com/guides/humidifier-field-guide",
  "title": "Humidifier field guide for shopping agents: questions to ask, evidence to require, when none fit",
  "description": "Ask about room size, noise and runtime; require values with units and test conditions; separate unknown from incompatible; say none fit when no humidifier meets a hard requirement. A neutral rubric with red flags and a worked example. No product picks.",
  "answer": "Before comparing humidifiers, ask the shopper for the room size, where and when the unit must be quiet, and how long it must run between refills. Accept only values with units and the setting or test condition they apply to, because an unqualified \"quiet\" or \"large room\" cannot verify a hard requirement. If every candidate fails a stated requirement, say none fit; if the evidence is missing, say what cannot be verified and ask for it.",
  "take": "Ask before you guess. A shopper who says \"quiet\" may mean a bedroom at night or an office at noon, so ask for the room size and refill tolerance, then compare only listings that state numbers.",
  "summary": "Ask the question the product page did not answer. Do not fill the gap with a guess.",
  "claims": [
    {
      "text": "For the search \"humidifier for dry winter air large room quiet top fill\", 4 of the 10 product descriptions returned printed a decibel figure, and none of those 4 stated the setting, distance or test method. Counting Shopify-generated metadata as well, 6 of the 10 records carried a decibel figure; no title did. Four descriptions stated a room area, none with its test conditions.",
      "source_title": "Cart for Agents: Receipts cited by our guides (Shopify agent-catalog snapshot, Oct 26, 2026)",
      "source_url": "https://cartforagents.com/research/agent-shopping-research#receipts-cited-by-our-guides",
      "source_date": "2026-10-26",
      "method": "observed",
      "limitation": "One search, one unpersonalized snapshot. The 4-of-10 counts are the description field; the 6-of-10 count adds Shopify-generated metadata. Describes what was written, not how loud or capable any unit is."
    },
    {
      "text": "In the same snapshot, one description said \"up to 32 hours\" while the Shopify-generated specification list in that record said \"Up to 36 hours\", and another generated list gave a 150 m² room area that does not appear in the description received.",
      "source_title": "Cart for Agents: Receipts cited by our guides (Shopify agent-catalog snapshot, Oct 26, 2026)",
      "source_url": "https://cartforagents.com/research/agent-shopping-research#receipts-cited-by-our-guides",
      "source_date": "2026-10-26",
      "method": "observed",
      "limitation": "We cannot see Shopify's inputs; a generated value may come from data we did not receive. Two examples, not an error rate."
    },
    {
      "text": "One description stated a 6.2 L tank, 320 mL/h output and up to 62 hours of runtime. At 320 mL/h the tank lasts about 19 hours, so the 62-hour figure implies a lower setting averaging about 100 mL/h.",
      "source_title": "Cart for Agents: Receipts cited by our guides (Shopify agent-catalog snapshot, Oct 26, 2026)",
      "source_url": "https://cartforagents.com/research/agent-shopping-research#receipts-cited-by-our-guides",
      "source_date": "2026-10-26",
      "method": "inferred",
      "limitation": "Arithmetic on stated figures only; real output varies with setting and conditions. We did not test the unit."
    },
    {
      "text": "Shopify says some Global Catalog fields, including the description, technical specifications and top features, are generated or enriched by Shopify, marked Inferred, may vary in accuracy, and should not be treated as merchant-authored source text.",
      "source_title": "Shopify: Global Catalog MCP, Inferred response fields",
      "source_url": "https://shopify.dev/docs/agents/catalog/global-catalog",
      "source_date": "2026-10-27",
      "method": "documented",
      "limitation": "Living developer documentation checked on this date. It does not say which values are generated for a given product."
    },
    {
      "text": "OpenAI says ChatGPT's product labels are generated and are not guarantees, that reviews and ratings are not verified by OpenAI, and recommends that shoppers verify products meet their needs before buying.",
      "source_title": "OpenAI Help Center: Shopping with ChatGPT Search",
      "source_url": "https://help.openai.com/en/articles/11128490-shopping-with-chatgpt-search",
      "source_date": "2026-10-27",
      "method": "documented",
      "limitation": "Living help article checked on this date; describes one assistant."
    },
    {
      "text": "Our earlier supplied-snippet simulation required each model to pick two of eight listings and offered no \"none of these fit\" option, so it did not test whether a model correctly rejects every candidate.",
      "source_title": "Cart for Agents: Vague vs specific product facts, simulated picks by Claude, GPT and Gemini models",
      "source_url": "https://cartforagents.com/research/vague-specific-product-facts-simulation",
      "source_date": "2026-10-27",
      "method": "controlled-simulation",
      "limitation": "Describes the design of our own simulation, which used invented products, supplied snippets and API models. The test of this guide described below is proposed and has not been run."
    },
    {
      "text": "In that simulation an invented humidifier sat among 7 real catalog listings, so picking 2 of 8 at random is 25%. With vague copy it was picked in 3 of 180 sessions across three Claude models, 0 of 60 on GPT-6.1-Sol, 6 of 60 on Gemini 3.8 Flash and 3 of 54 on Gemini 3.1 Pro. With its facts as a spec line: 134 of 180, 20 of 60, 19 of 59 and 15 of 55. On GPT and Gemini it was picked in every session with a detailed shopper brief (20 of 20, 19 of 19, 15 of 15) and in 0 of 120 sessions across those three models (0 of 40 each) with a one-line ask or no shopper. The real competing listings already stated 21, 26 or 28 dB.",
      "source_title": "Cart for Agents: Vague vs specific product facts, simulated picks by Claude, GPT and Gemini models",
      "source_url": "https://cartforagents.com/research/vague-specific-product-facts-simulation",
      "source_date": "2026-10-27",
      "method": "controlled-simulation",
      "limitation": "Invented product, supplied snippets, no page opening or tools, forced choice of two, a detailed shopper brief written to fit the test product, API models rather than the ChatGPT or Gemini shopping products. Not live visibility, rank or sales."
    }
  ],
  "sources": [
    {
      "title": "Cart for Agents: Receipts cited by our guides",
      "url": "https://cartforagents.com/research/agent-shopping-research#receipts-cited-by-our-guides",
      "accessed": "2026-10-27"
    },
    {
      "title": "Shopify: Global Catalog MCP",
      "url": "https://shopify.dev/docs/agents/catalog/global-catalog",
      "accessed": "2026-10-27"
    },
    {
      "title": "OpenAI Help Center: Shopping with ChatGPT Search",
      "url": "https://help.openai.com/en/articles/11128490-shopping-with-chatgpt-search",
      "accessed": "2026-10-27"
    },
    {
      "title": "Cart for Agents: Vague vs specific product facts, simulated picks by Claude, GPT and Gemini models",
      "url": "https://cartforagents.com/research/vague-specific-product-facts-simulation",
      "accessed": "2026-10-27"
    }
  ],
  "limitations": [
    "This guide has not been tested with agents or people. The usability test described here is a proposal.",
    "It is reference material for evaluating evidence. It does not recommend any product or brand and does not override a shopper's instructions or an assistant's own rules.",
    "It is not a safety, medical or acoustic standard. It does not establish that any humidifier is safe for a health condition, and it does not name a correct test method for noise or coverage.",
    "The catalog observations come from one search we ran in Shopify's agent catalog on October 26, 2026, without personalization. The worked example uses invented records.",
    "Cart for Agents offers merchants a free emailed product check. That service is separate from this rubric and does not influence it."
  ],
  "related": [
    "https://cartforagents.com/guides/spec-sold-product-guides",
    "https://cartforagents.com/guides/agent-ready-product-data",
    "https://cartforagents.com/research/quiet-humidifiers",
    "https://cartforagents.com/research/agent-shopping-research",
    "https://cartforagents.com/research/vague-specific-product-facts-simulation"
  ],
  "published": "2026-10-27",
  "updated": "2026-10-27",
  "checked": "2026-10-27",
  "kind": "field-guide"
}