MCP SERVER

Your AI visibility data — where your AI tools already are.

The Rankwize MCP Server connects your citation data, content library, and effectiveness scores to any MCP-compatible client — Claude Desktop, custom agents, or internal tooling. Pull current prompt status, topic effectiveness, and recent run summaries without leaving your workflow.

Included on Pro and above — Track plans expose monitoring data; +Diagnose adds diagnosis, recommendations, and topic effectiveness.

Seven tools, read-heavy by design.

Five tools are read-only. The two write tools — create_prompt and trigger_run — require the write scope on your API key. Read-only access is the default when you generate a key.

Track plans expose the monitoring tools — prompt status, runs, citations, and your content library. +Diagnose plans additionally unlock the diagnosis tools: fix types, recommendations, and topic effectiveness.

list_prompts

Tracked intents with current citation status per platform

get_prompt_detail

Factor scores, citation history, matched page, current recommendation (diagnosis fields require +Diagnose)

get_content_library

Indexed content pages with fingerprint data

get_topic_effectiveness

Effectiveness scores by topic, fix type, and platform (+Diagnose)

get_run_summary

Results from recent monitoring batch runs

create_prompt

Add a new prompt intent to your tracking queue

trigger_run

Enqueue an on-demand monitoring batch

list_prompts → response

GET /mcp
Tool: list_prompts
Authorization: Bearer rk_live_xxxx

→ {
  "prompts": [
    {
      "id": "pi_01jxxxxxxx",
      "intent": "best ai visibility tool for b2b saas",
      "platforms": {
        "chatgpt": "cited",
        "perplexity": "invisible",
        "google_ai_overviews": "mentioned"
      },
      "priority_score": 82,
      "fix_type": "Add_Evidence"
    }
  ],
  "total": 94
}

Generate a key

Settings → API Keys → New key. Assign read or write scope. Label it so you know which agent or workflow is using it. Revoke at any time.

Connect your client

Add the Rankwize MCP server URL to your client config. Pass your key as a Bearer token. No OAuth. No shared credentials.

HTTP/SSE transport. Per-tenant API key auth.

The server uses HTTP with SSE for streaming responses. Authentication is per-tenant API key — generate keys in Settings, assign read or write scope, revoke at any time. No OAuth flow. No shared credentials.

Each key is labeled so you can track which agent or workflow is using which key. Rotate without downtime by generating a replacement before revoking the old one.

What you can build.

read scope

An agent that surfaces your lowest-performing prompts in a daily standup summary
A dashboard that pulls current fix type distribution without opening Rankwize
A Slack notification when a high-priority prompt drops citation

write scope

Queue new prompt intents from keyword research output automatically
Trigger a monitoring batch after your content team completes a production run
Build a prompt-lifecycle workflow without touching the Rankwize UI

The MCP Server doesn't push data to you. It makes your data available to pull when your workflow needs it.

How it works

Generate a key in Settings

Create an API key, assign read or write scope, and label it for the workflow using it.

Add the server URL to your MCP client config

Compatible with Claude Desktop, Cursor, n8n, and any client that supports the MCP protocol.

Call any of the 7 tools to pull current data

Prompt lists, effectiveness scores, run summaries — available to your agent or workflow on demand.

Included on Pro and above.

The MCP Server is included on Pro and above — across both Track and Diagnose plans, with no separate purchase. Track plans expose the monitoring tools; +Diagnose plans add the diagnosis tools.

Monthly MCP units are included by plan: 10k on Pro Track, 50k on Business Track, 150k on Scale Track, 25k on Pro + Diagnose, 75k on Business + Diagnose, and 300k on Scale + Diagnose. Per-minute limits protect against runaway scripts.

Track plans — monitoring tools

Query prompt status, monitoring runs, citation and mention visibility, and your content library. Create prompts and trigger allowed monitoring runs from connected tools.

+Diagnose — diagnosis tools

Everything in the monitoring set, plus gap diagnosis, fix types, recommendations, and topic effectiveness. The two write tools (create_prompt, trigger_run) require explicit per-key opt-in, so a read-only key can never mutate your workspace.

Included on Pro and above — Track and Diagnose.