Give your agent the keys
to your SEO data
One URL connects Claude, Codex, Gemini or any MCP client to your Ranksmile workspace.
See what it can doclaude mcp add --transport http ranksmile https://app.ranksmile.pl/mcpWhat your agent can ask for
Every tool reads. Nothing writes, nothing publishes, nothing deletes, and each call re-derives what the connected account is allowed to see.
Workspaces and sites
Where everything hangs off: the brands, domains and projects on the account.
workspace__listEvery workspace the connected account can reach, with its primary domain.
site__getOne site: its domain, its tracked countries, and when it was last crawled.
Articles and content scores
The editor side. This is where an agent finds out why a draft scores the way it does.
article__listArticles with their stored scores and word counts. Filter by workspace or status.
article__getOne article: metadata, stored scores, and optionally the HTML body.
article__scoreThe per-slot breakdown: word, heading and paragraph counts against competitor-derived targets, and every NLP term with its current and required usage.
article__optimize_logAuto-Optimize history: score and length before and after, the mode, the phase, and the rejection reason when a run changed nothing.
article__jobsRecent generation jobs with status, payload and result, so you can see what the generator was told and what it returned.
AI visibility
What the answer engines said, stored whole, with the sources they cited and the brands they named.
visibility__scoreThe Visibility Score per engine: the share of tracked prompts where the brand is named, with the previous value beside it.
prompt__listTracked prompts by topic, with their priority tier and last scan time.
prompt__answerOne stored answer: the text an engine returned, the URLs it cited, the brands it named with sentiment, and the fanout queries it expanded into.
competitor__listThe brands named beside you or instead of you, ranked by how often.
source__listThe URLs the engines lean on for your queries, and which brands each source mentions.
Rank tracking
Classic Google positions on the schedule you set, desktop and mobile, by country.
keyword__listTracked keywords with current position, previous position and the delta.
keyword__historyThe position history for one keyword, per device and per country.
search_console__performanceImpressions, clicks and average position from Search Console for a page or a query.
Site audit
What the last crawl found.
audit__issuesSite Audit findings for a crawl, grouped by severity.
Stop copying numbers between tabs. Let the agent read them
Ask why, not just what
Your agent reads the score breakdown, not the number: every NLP term, every count against its target.
See the Content ScoreWork without the app
Pull rankings, AI answers and drafts into whatever you already work in.
See AI VisibilityRead-only by design
Every tool reads. Nothing writes, nothing publishes, nothing deletes.
View example
Connect it in under a minute
Copy the endpoint from Settings, paste it into your agent, approve the consent screen. That is the whole setup.
Start your free trial







