Model Context Protocol · Beta

Bring rank.ai into your AI stack.

Run audits, query reports, and ask follow-up questions about the data — all from Claude, Cursor, ChatGPT, or any MCP-aware AI assistant. No SDK install, just a key and a config block.

Conversational audits

Drop a domain into Claude and watch it scrape locations, propose keywords, ask you to confirm, charge credits, and ship a report — all from chat.

Talk to your data

Ask 'where am I ranking worst?' or 'who's beating me on cosmetic dentistry?' Histograms by default, raw points only when you drill in.

Works with any MCP host

Claude Desktop, Cursor, Claude Code, and any other MCP-aware client. One API key, one config block, no SDK install.

One config block.

Paste this into your MCP host's config and you're connected. The example is for Claude Desktop's claude_desktop_config.json; every other host is one or two field-name swaps away.

{
  "mcpServers": {
    "rank-ai": {
      "url": "https://api.rank.ai/mcp",
      "headers": {
        "Authorization": "Bearer rnk_live_..."
      }
    }
  }
}

Generate the API key at your account settings. Each key is shown exactly once — copy it before you close the dialog.

Tools at a glance.

Four tool families: audit.* + report.* for local SEO audits, national_rank.* for SERP tracking, and ai_rank.* for ChatGPT / Claude / Gemini visibility.

audit.*

  • audit.create — start with a domain
  • audit.next — what to do next
  • audit.answer — confirm or edit
  • audit.cancel — kill in-flight
  • audit.usage — credit balance
  • audit.list — your sessions

report.*

  • report.summary — top KPIs
  • report.list_keywords
  • report.list_locations
  • report.location — single location
  • report.grid_summary — histograms
  • report.grid_points — drill-down
  • report.competitor
  • report.findings
  • report.insights — pre-computed callouts

national_rank.*

Daily Google SERP position monitoring per keyword.

  • national_rank.list_keywords — tracked keywords + latest rank
  • national_rank.keyword_history — time-series of rank checks

ai_rank.*

Visibility in ChatGPT, Claude & Gemini answers — citations, mentions, share-of-voice, underlying search queries.

  • ai_rank.list_prompts — tracked prompts + per-provider visibility
  • ai_rank.create_prompt — track a new prompt
  • ai_rank.run_now — ad-hoc check (charges credits)
  • ai_rank.get_run_detail — answer text + queries + citations
  • ai_rank.suggest_prompts — auto-generate from your keywords

Example agent calls: "What's our visibility trend in ChatGPT this week?" · "Track this keyword in AI Rank too" · "Show me the underlying searches Claude ran for our 'best Botox' prompt."

Full reference + token-cost details in the docs.

Ready to Improve

Your Rankings?

Use our free tools to get instant insights into your SEO performance and discover opportunities to rank higher