Skip to main content
DevTune exposes a Model Context Protocol (MCP) server so that AI coding agents can query your AI search visibility data directly. This lets agents like Claude, Cursor, and other MCP-compatible tools access your metrics without leaving the IDE.

​
Transport

The server uses Streamable HTTP transport at: 
https://app.devtune.ai/api/mcp
Authentication is the same as the REST API — pass your API key in the Authorization header: 
Authorization: Bearer dtk_live_your_key_here

​
Available Tools

ToolDescription
devtune_get_visibility_summaryShare of voice, presence rate, brand awareness, and citation counts
devtune_get_competitive_positionCompetitive positioning time series (daily/weekly/monthly)
devtune_get_citation_analysisPaginated citations with URL, domain, classification, and position
devtune_get_actionsCurrent actions workspace, including active recommendations and adopted backlog items
devtune_get_action_briefStored brief status and execution-ready markdown for a specific action
devtune_get_traffic_summaryLLM bot and referral traffic analytics
devtune_get_adoption_metricsnpm downloads and GitHub stars trends
devtune_get_content_gapsContent gaps where competitors win and you are missing
Each tool accepts optional filter parameters (dates, platforms, pagination) and returns JSON data matching the corresponding REST API endpoint. All MCP tools are read-only. MCP tools honor the same API key scope restrictions as the REST API. For example, devtune_get_actions requires actions.read, and devtune_get_content_gaps requires intelligence.read.
Note: The devtune_get_content_gaps tool is exclusive to the MCP server and has no REST API equivalent. All other tools map directly to a REST endpoint.
For devtune_get_actions, the most useful filters are:
  • surface: recommendation or backlog
  • status: active, backlog, in_progress, blocked, done, canceled
  • aliases: open, completed, and dismissed map to backlog, done, and canceled
  • priority and channel for narrower work queues

​
Setup

​
Claude Desktop

Add to your Claude Desktop config file: macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json 
{
  "mcpServers": {
    "devtune": {
      "type": "streamable-http",
      "url": "https://app.devtune.ai/api/mcp",
      "headers": {
        "Authorization": "Bearer dtk_live_YOUR_API_KEY_HERE"
      }
    }
  }
}

​
Cursor

Open Settings > MCP Servers > Add Server, or edit .cursor/mcp.json: 
{
  "mcpServers": {
    "devtune": {
      "type": "streamable-http",
      "url": "https://app.devtune.ai/api/mcp",
      "headers": {
        "Authorization": "Bearer dtk_live_YOUR_API_KEY_HERE"
      }
    }
  }
}

​
Claude Code

claude mcp add devtune \
  --transport http \
  https://app.devtune.ai/api/mcp \
  --header "Authorization: Bearer dtk_live_YOUR_API_KEY_HERE"

​
Example Prompts

Once connected, you can ask your AI agent questions like:
  • “What is my current share of voice across AI platforms?”
  • “Show me the competitive position trend for the last 30 days”
  • “List active recommendations for this project”
  • “Load the stored brief for action 6f1e2d3c-1a2b-4c5d-8e9f-0123456789ab and summarize the execution plan”
  • “Show backlog items that are blocked or in progress”
  • “What content gaps do my competitors have that I’m missing?”
  • “How much traffic am I getting from AI chatbots?”

​
How It Works

The MCP server reuses the same data-fetching layer as the REST API. Your API key determines which project the tools operate on, and all requests go through the same rate limits and authentication. The server runs in stateless mode — each request is independent with no session persistence. This is the recommended mode for HTTP-based MCP servers.