Яндекс.Метрика

TamTam: is the website ready for AI agents

tamtam.chat·checked November 11, 2026
39out of 100
Critical

Agents struggle to read the site. Start with the first items on the list.

Paste the task into Claude Code, Cursor or another agent with access to the site’s code.

  • Agent access: 59
  • Discoverability: 27
  • Content for models: 14
  • AI bot rules: 50
  • API and protocols: —

Fix these first

First, what gets in agents’ way the most.

  1. 01

    Text is visible without JavaScript

    Failed

    Render homepage text on the server: SSR or a static build. The HTML must contain at least 300 words of visible text without running JavaScript.

    Most agents do not run JavaScript. Product text has to be in the page itself, not loaded by a script.

    Access without CAPTCHA or JS →
    ›What the site returned
    URL
    https://tamtam.chat/
    Found
    words of text: 0

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/"
  2. 02

    The site serves Markdown when an agent asks

    Failed

    When the homepage is requested with “Accept: text/markdown”, return Markdown with “Content-Type: text/markdown” and “Vary: Accept”. Keep serving HTML for ordinary requests.

    An agent can ask for a page in Markdown instead of HTML and spend far less of its budget on markup.

    Markdown for agents →
    ›What the site returned
    URL
    https://tamtam.chat/
    Response code
    200
    Headers
    content-type: text/html; charset=utf-8

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -H "Accept: text/markdown" -D - -o /dev/null "https://tamtam.chat/"
  3. 03

    Prices are visible without JavaScript

    Failed

    Link to the pricing page from the homepage and put the prices in that page’s HTML, not in a script that fills them in after load.

    Agents are often asked to compare pricing. If a script fills in the prices, the agent never sees them.

    Access without CAPTCHA or JS →
    ›What the site returned
    URL
    https://tamtam.chat/

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/"
  4. 04

    A sitemap exists

    Failed

    Generate a sitemap.xml listing your pages and reference it in robots.txt with a “Sitemap: https://your-site/sitemap.xml” line.

    sitemap.xml is the list of pages. An agent uses it to find sections it did not reach by links.

    sitemap and JSON-LD →
    ›What the site returned
    URL
    https://tamtam.chat/sitemap-index-tamtam-chatchannel.xml.gz
    Response code
    404

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/sitemap-index-tamtam-chatchannel.xml.gz"
  5. 05

    Company or product markup is present

    Failed

    Add JSON-LD markup of type Organization or SoftwareApplication with name and url to the homepage.

    JSON-LD is company data in a machine format: name, site URL, product type.

    sitemap and JSON-LD →
    ›What the site returned
    URL
    https://tamtam.chat/

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/"
Other issues15
  1. 06

    llms.txt exists

    Failed

    Put an llms.txt in Markdown at the site root: a top-level heading with the product name, a short description and a list of links to key pages. The links must resolve.

    A short file for language models: what the product is and where the key pages are.

    llms.txt →
    ›What the site returned
    URL
    https://tamtam.chat/llms.txt
    Response code
    200
    Headers
    content-type: text/html; charset=utf-8

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -H "Accept: text/markdown" -D - -o /dev/null "https://tamtam.chat/llms.txt"
  2. 07

    Pages have Markdown versions

    Partial

    Serve a Markdown version of pages at the URL ending in .md and reference it with <link rel="alternate" type="text/markdown"> on the page itself.

    The same page at a URL ending in .md: clean text with no menus or scripts.

    Markdown for agents →
    ›What the site returned
    URL
    https://tamtam.chat/index.md
    Response code
    200

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -H "Accept: text/markdown" -D - -o /dev/null "https://tamtam.chat/index.md"
  3. 08

    A clear heading structure

    Failed

    Keep a single h1 on the homepage, add h2 headings, do not skip levels (h3 follows h2, not h4) and wrap the main content in a <main> tag.

    One main heading, subheadings in order and a marked main area. An agent uses them to split the page into parts.

    Access without CAPTCHA or JS →
    ›What the site returned
    URL
    https://tamtam.chat/
    Found
    H1 headings: 0
    H2 headings: 0
    has <main>: no

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/"
  4. 09

    robots.txt has rules for AI bots

    Failed

    Add separate User-agent groups for AI bots to robots.txt: GPTBot, ClaudeBot, OAI-SearchBot, PerplexityBot, YandexAdditional. Naming two is enough.

    AI bots are listed by name. Whether you allow or block them, it shows the choice was made.

    robots.txt for AI →
    ›What the site returned
    URL
    https://tamtam.chat/robots.txt
    Response code
    200

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/robots.txt"
  5. 10

    Alice AI robots are allowed

    Partial

    Add a group to robots.txt with “User-agent: YandexAdditional” and “User-agent: YandexAdditionalBot” and an “Allow: /” rule.

    The YandexAdditional robots bring a site into Alice AI answers. Best to allow them with a rule of their own.

    robots.txt for AI →
    ›What the site returned
    URL
    https://tamtam.chat/robots.txt
    Response code
    200

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/robots.txt"
  6. 11

    A missing page returns 404

    Failed

    Configure the server so that a missing URL returns 404, not 200 and not a redirect to the homepage.

    If the site answers “OK” to any URL, the agent takes an empty page for a real one.

    Access without CAPTCHA or JS →
    ›What the site returned
    URL
    https://tamtam.chat/
    Response code
    200
    Headers
    content-type: text/html; charset=utf-8

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/"
  7. 12

    The homepage links to API docs

    Failed

    Link to your API docs from the homepage, for example in the footer or a developers menu.

    When asked to set up an integration, an agent looks for docs starting from the homepage.

    sitemap and JSON-LD →
    ›What the site returned
    URL
    https://tamtam.chat/

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/"
  8. 13

    The site points agents to its machine-readable files

    Failed

    Add a Link header or a <link> tag on the homepage pointing to llms.txt, the page’s Markdown version (rel="alternate" type="text/markdown") or the API description (rel="service-desc").

    Links in the response header or page code lead to llms.txt, a Markdown version or an API description.

    Markdown for agents →
    ›What the site returned
    URL
    https://tamtam.chat/

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/"
  9. 14

    llms-full.txt exists

    Failed

    Put an llms-full.txt at the site root with the full text of the site or docs in Markdown.

    The full text of the site or docs in one file. An agent reads it in a single request.

    llms.txt →
    ›What the site returned
    URL
    https://tamtam.chat/llms-full.txt
    Response code
    200
    Headers
    content-type: text/html; charset=utf-8

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -H "Accept: text/markdown" -D - -o /dev/null "https://tamtam.chat/llms-full.txt"
  10. 15

    Content use is declared

    Failed

    Add a “Content-Signal: search=yes, ai-input=yes, ai-train=yes” line to robots.txt. Choose yes or no for each value yourself.

    The Content-Signal line says whether your text may be used for search, AI answers and model training.

    robots.txt for AI →
    ›What the site returned
    URL
    https://tamtam.chat/robots.txt
    Response code
    200

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/robots.txt"
  11. 16

    Company markup has a description and contacts

    Failed

    Add description, logo and contactPoint (address, email or telephone also work) to your Organization JSON-LD.

    The description, logo and contacts in the markup let an assistant confirm this is a real company before recommending it.

    sitemap and JSON-LD →
    ›What the site returned
    URL
    https://tamtam.chat/

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/"
  12. 17

    About, contact and policy pages are linked from the homepage

    Failed

    Link to the About, Contact and privacy policy pages from the homepage, for example in the footer.

    An agent checks these pages to make sure the site can be trusted.

    sitemap and JSON-LD →
    ›What the site returned
    URL
    https://tamtam.chat/

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/"
  13. 18

    Pricing is available in Markdown

    Failed

    Serve the pricing page in Markdown: at the page URL ending in .md or at /pricing.md. The file must contain prices.

    An agent comparing products reads pricing most easily from a plain text file.

    Markdown for agents →
    ›What the site returned
    URL
    https://tamtam.chat/pricing.md
    Response code
    200
    Headers
    content-type: text/html; charset=utf-8

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/pricing.md"
  14. 19

    Homepage metadata is filled in

    Partial

    Add title, meta description and link rel="canonical" to the homepage <head>, and a lang attribute to the html tag.

    Title, description, canonical URL and language. They tell an agent what the site is.

    sitemap and JSON-LD →
    ›What the site returned
    URL
    https://tamtam.chat/

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/"
  15. 20

    Markup links to official profiles

    Failed

    Add a sameAs field to your JSON-LD with links to official profiles: GitHub, social networks, app stores.

    sameAs links to GitHub, social networks and app stores. They tell an assistant that mentions elsewhere are about you.

    sitemap and JSON-LD →
    ›What the site returned
    URL
    https://tamtam.chat/
    Found
    sameAs links: 0

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/"
What already works8

  • Homepage opens without a CAPTCHA

    ›What the site returned
    URL
    https://tamtam.chat/
    Response code
    200
    Headers
    content-type: text/html; charset=utf-8

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/"

  • A short path from http to https

    ›What the site returned
    URL
    https://tamtam.chat/
    Response code
    200
    Headers
    content-type: text/html; charset=utf-8
    Found
    redirects: 1

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/"

  • The homepage is light and fast

    ›What the site returned
    URL
    https://tamtam.chat/
    Found
    page size, KB: 15
    response time, s: 0.454

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/"

  • robots.txt exists

    ›What the site returned
    URL
    https://tamtam.chat/robots.txt
    Response code
    200
    Headers
    content-type: text/plain; charset=utf-8

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/robots.txt"

  • Agents and AI search are allowed

    ›What the site returned
    URL
    https://tamtam.chat/robots.txt
    Response code
    200

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/robots.txt"

  • The homepage is not hidden from search

    ›What the site returned
    URL
    https://tamtam.chat/

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/"

  • ChatGPT and Claude get the same page people do

    ›What the site returned
    URL
    https://tamtam.chat/
    Response code
    200
    Headers
    content-type: text/html; charset=utf-8

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/"

  • Page preview data is filled in

    ›What the site returned
    URL
    https://tamtam.chat/
    Found
    found: og:title, og:description, og:image, og:type

    Repeat the check yourself

    curl -sSL -c /dev/null -A "agent-readiness-index/0.1" -D - -o /dev/null "https://tamtam.chat/"
Not scored18

These checks do not apply to the site and do not lower the score.

  • Open API docs or an MCP server exist
  • An API description (OpenAPI) is published
  • An MCP server exists
  • An agent can find how to sign in
  • The API reports errors in a machine format
  • An API catalog exists
  • Agent skills are published
  • The API says which server grants access
  • Homepage text fits an agent’s memory
  • Content starts near the top of the page
  • llms.txt is structured well enough to navigate
  • The Markdown version has a header with page data
  • The MCP server answers a connection
  • MCP tools have descriptions
  • An access refusal says where to sign in
  • An agent-to-agent card exists
  • There is a guide for how an agent gets access
  • A catalog of AI interfaces exists

Check your own site

Paste the address and get the same breakdown in half a minute.

Check a site →

Other checked sites