Skip to main content
Site Health is the page for a project’s website. Open Site Health in the sidebar, under Site. It brings three checks together. Each has its own score and its own page.
  • Site Readiness checks whether search engines can crawl, read and index your pages, and whether the AI search products built on them can reach you.
  • Agent Readiness checks whether an AI agent working in a web browser can open your pages and use them.
  • AI Indexing counts the pages of yours that AI platforms cite in their answers.
Content Audit is a separate page. It scores your content. Site Health checks the site the content lives on.

​
The dashboard

Reading from the top:
  • Latest results. One tile for each check, with its score, what the score is made of, and how it moved since the previous check. Under the Site Readiness and Agent Readiness scores, a thin bar splits the pages by state: ready, needs work, blocked and not checked. Under AI Indexing, it shows which platforms cite you. Select a tile to open that check. While a check runs, its tile shows the progress next to the last result.
  • Next steps. What to do first, across all three checks. Two rows show, and Show N more opens the rest. Review opens the check with that item expanded. Items are ordered by how much they hold you back: blocked work that keeps a search crawler out, other blocked work, things that got worse since the last check, then problems by the number of pages they affect, then setup such as Turn on weekly checks.
  • Over time. Site Readiness and Agent Readiness over your last 12 checks, with each line’s current value and change. Trend lines start after your second check. A check measured with an older method starts a new line, so the chart never joins two methods.
  • Pages. The pages that matter most, with their Site and Agent state and how many AI platforms cite each. Blocked pages that AI still cites come first, then the most-cited pages, then the homepage. All N pages opens the full list.
  • AI crawler access. Whether the search crawlers behind AI answers can reach your site, and how they are blocked when they are. Training crawlers get one quiet line.
  • AI platforms. How many of your pages each platform cited in the last 30 days.
  • Recent checks. The latest four. All checks lists every check and filters by Site Readiness, Agent Readiness or AI Indexing probes.

​
How the scores work

Every score is measured from your site. There is no overall grade and no weighted blend, and OpenLens does not show a number it did not measure. A few things hold for all three:
  • Pages we couldn’t measure show as Not checked. They are listed, but they aren’t counted, so a page we couldn’t reach never reads as passing or failing.
  • Percentages round down, so 100% means every page we measured was ready. When it matters, read the count under the score, such as “199 of 200 pages search-ready”, which is 99%.
  • A new method starts a new line. When we change what a check measures, the first check afterwards says “First check with the updated method” and isn’t compared with older ones.
  • Agent Readiness needs enough pages. If we could measure fewer than half the pages it sampled, it shows no score. It says “Not enough pages measured”, and why.
  • AI Indexing is a count on purpose. Pages AI cites are often ones no check has covered yet, so a percentage would measure which pages we happened to check. See AI Indexing.

​
Run checks

Run checks starts a Site Readiness check and an Agent Readiness check together, on your saved scope. They are two checks, and each has its own page. If your Agent Readiness allowance for the month is used up, only the Site Readiness check starts. You stay on the dashboard, and each tile shows its progress with a link to watch it live. AI Indexing doesn’t need a check. It updates every time a Prompt Tracking run finishes. Probes, which ask AI search whether a specific page comes back, start from the AI Indexing page.

​
Set check scope

Set check scope chooses Pages per check (the choices depend on your plan) and Only pages under, a path such as /guides. Agent Readiness opens a few of those pages in a browser: 3 on Free, 10 on Starter and 50 on Agency. The scope you set is saved in your browser. Another browser, or a teammate, starts from the defaults.

​
Schedules

Open Schedules from the ⋯ menu. There is one row each for Site Readiness, Agent Readiness and AI Indexing probes. Each row is Off, Weekly or Monthly (Daily on Agency), at an hour in UTC.
  • Free: no schedules.
  • Starter: weekly or monthly.
  • Agency: daily, weekly or monthly.
A scheduled check doesn’t use the scope you set above. Site Readiness checks up to 500 pages on Starter and 40,000 on Agency. Agent Readiness opens your plan’s sample. If your monthly page allowance is used up, the scheduled check doesn’t run. Prompt runs keep their own schedule in Settings.

​
Open a check

Every check has its own page, built the same way as a Prompt Tracking run.
  • The header. The feature, your domain, and a line with the check’s name, date and number of pages, and whether it was manual or scheduled. Checks are named by feature: “Site check 14”, “Agent check 6”, “Probe 7”. Rename in the ⋯ menu lets you call one something else.
  • While it runs. A bar shows the step and the count, such as “Checking pages · 25 of 50”, with the time elapsed and Cancel. Results fill in as pages are checked.
  • Four tiles with the check’s headline numbers.
  • To do. What to fix, most important first. Select a row to expand it: what we found, what we measured, the affected pages, and the steps. Where your plan includes it, a fix can come with a snippet to copy, and Recheck checks the fix against your site.
  • Pages. Every page in the check. Filter by state, and select a page to see its issues and facts.
  • What we checked. One line per check with its outcome, beside a card for the feature: AI crawler access, Browser access, or the prompts behind your citations.
If a check didn’t finish, its page says so, gives the reason, and links to the last check that still stands. Download PDF and Copy share link in the ⋯ menu are for Site Readiness checks, on Starter and Agency.

​
Plans and roles

Limits and monthly allowances are in Limits and allowances. Project viewers can read everything on the page. They can copy a snippet and download a PDF when the plan includes one. They don’t see Run checks, Set check scope, Schedules, Recheck or an Upgrade button, and the dashboard says View only. Client logins don’t have Site Health in their sidebar.

​
What OpenLens does not change

OpenLens never changes your site. Site Health gives you the steps and the snippet. You apply them, then Recheck shows whether the fix worked.

​
Google data

Site Health doesn’t show Search Console or GA4 data. If your organization has connected Google in Settings → Google integrations, the in-app agent can read it. See Settings.

​
Through MCP

  • run_readiness_check starts a Site Readiness check.
  • get_readiness_run reads a Site Readiness check’s status and results.
  • list_readiness_runs lists a project’s Site Readiness checks.
  • run_ai_index_coverage, get_ai_index_coverage_run and list_ai_index_coverage_runs work with AI Indexing probes.
Agent Readiness checks aren’t available over MCP. The client role can’t start or view readiness checks through MCP. Links to the old Site page, to /dashboard/readiness, /dashboard/site-readiness and /dashboard/ai-indexing, and to the Google and Search Traffic pages open Site Health. A link that names a project opens that project’s page. A link that names only a check opens the check on its own project and feature. Any other link opens your first project.