# Methodology

> Source: AgentsIndex — https://agentsindex.ai/methodology
> Index last updated: 17 November 2026

What gets listed, where the facts come from and how listings stay current.

## The checks behind every listing

Each tool page is read off the tool's own site, then read again on a schedule.

- **421** tools in the index
- **96%** of 421 re-read on their own site this cycle
- **90 days** at most between two reads of one tool
- **17 Nov 2026** last time we read a tool's site

## Where the facts come from

Each listing starts from the vendor's own site: product pages, docs, pricing and changelog. We add GitHub activity for open-source tools, and benchmarks and community discussion where they exist. Every listing links to its sources.

## What we check

**The product exists.** The site loads, the app runs or the repo is active. **The price is current.** Taken from the vendor's pricing page on the date shown on the listing. **Features are real.** We look for them in the docs, not just the marketing copy.

Listings are written from these public sources, not from hands-on trials. We don't see private pricing or enterprise contracts.

## How tools get listed

The team behind a tool [submits it](https://agentsindex.ai/submit). If it fits the index, the team joins with a one-time payment and we research the tool before its page goes live. Own a tool that is already listed? Use **Claim** on its page and show that the product is yours.

## How tools are ordered

Featured listings sit in a labelled block at the top of their category and first on related pages. Below them, each category is ranked by what we can confirm about each tool: a public price, an API and docs, independent sources and stated limits. Below the Featured places, paying never moves a listing.

## How listings stay current

We re-read the oldest listings first and refresh anything we know has changed: a rename, an acquisition, a new product. Last checked across the index on 17 November 2026.

## Reading the yes, partly and no marks

- **Yes**: The tool's own docs or pricing page say it does this.
- **Partly**: It does, with a catch: a higher plan, an add-on, a beta or a usage cap.
- **No**: Its own pages say it does not.
- **Not stated**: Its pages don't say either way.

## How we count the figures

### In our articles

- Every number in an article we write comes from our own records of each tool or from a source the article cites, and is checked against them before the article goes out.
- Prices are read from each tool's own pricing page, or from the marketplace listing it sells on when its own site shows none, and each ranking shows the days we read them.
- When a pricing page offers monthly and yearly billing, we record the monthly price and keep the yearly one beside it. A plan sold only by the year is shown as billed yearly.
- The typical price is the middle value of each tool's cheapest paid plan, so free plans stay out of it. If fewer than 5 tools have a price, we give no typical price.
- Counts such as how many tools have a free plan use the same records and rules as our category pages, so an article and its category give the same answer.

### In our price tracker

- Once a week we read every tool's own pricing page. No AI model reads the page.
- A price change counts only when the same plan, in the same currency and billing period, shows the old price and then the new one, each in two readings at least a day apart.
- A move of more than 5 times, up or down, doesn't count: at that size it is almost always a different plan or unit, not a new price.
- Each change shows the price as the pricing page prints it, not converted or rounded.
- Our price changes page lists the last 13 weeks, newest first. Fixes to our own records are not price changes.

### In our AI answer checks

- Each month we put the questions buyers ask to ChatGPT and Perplexity, and keep every answer and the sources it cites.
- We don't ask Gemini, Google AI Overviews or Copilot, and no date is set to add them.

### In our study: AI Tool Pricing Index

- Only tools whose prices we checked in the last 60 days count. A median needs at least 5 of them, and a smaller group shows counts only.
- The starting price is the cheapest paid plan, so a free plan never counts in it. A free trial is not counted as a free plan.
- A yearly price counts at a twelfth of the year. Prices by usage, one-time prices and prices on request stay out of the medians.
- Only prices in US dollars count, and nothing is converted.
- Each month goes live on its own once every check passes. A month that fails one, such as too few prices, a sharp move or a change in method, waits until a person has reviewed it.

[Read the study](https://agentsindex.ai/data/ai-tool-pricing-index)

### In our study: What AI answers about AI tools cite

- Once a month we ask Perplexity and ChatGPT the same set of questions about AI tools and keep each answer with the sources it lists.
- This edition counts the 62 answers collected between 6 November 2026 and 7 November 2026, to 31 questions.
- A site counts once per answer that lists at least one of its pages, and a site is its main domain, so en.wikipedia.org is wikipedia.org.
- Shares are of the 62 answers that listed any source.
- A product site is the website of a product listed here.

[Read the study](https://agentsindex.ai/data/what-ai-cites-on-ai-tools)

## Who runs AgentsIndex

AgentsIndex is written and run from the Netherlands by Team AgentsIndex. Every listing, guide and report is published under the site's own name, and the team answers for all of it.

## Changes to a listing

Changes a maker asks for are made only on a paid listing. A listing that has never been paid for shows what our own data says.

Anyone can report a wrong fact with “Report an error” on a listing, at no cost. We fix it once the tool's own website confirms it. Paying buys changes and speed, never accuracy.

## Where we use AI

AI models do the reading and write the drafts. A model reads each tool's own website and drafts its listing. Our articles and guides are drafted by a model from our own data.

No person reviews our drafts before they go live; our checks decide. Every fact on a listing must point to its source on the tool's website, and a sentence that website does not support is cut. In an article, every figure must match our data and every quote must appear word for word in its source. An article that fails a check is not published.
