MCP Server

Fload provides an MCP (Model Context Protocol) server that lets AI clients like Claude Desktop, Cursor, or any MCP-compatible tool connect directly to your Fload data and actions.

What is MCP?

Model Context Protocol is an open standard that lets AI applications connect to external tools and data sources. Instead of copy-pasting data into your AI chat, MCP lets the AI pull data directly from Fload — and take actions on your behalf.

With the Fload MCP server, you can:

• Ask Claude Desktop about your app metrics and get live answers
• Have Cursor reference your app's reviews while writing code
• Build custom AI workflows that read and act on your Fload data

Setup

1. Get an API key

Generate an API key from Settings → API Keys in your Fload dashboard. See API Keys for details on creating and managing keys.

2. Configure your AI client

Claude Desktop — add this to your claude_desktop_config.json:

{
  "mcpServers": {
    "fload": {
      "command": "npx",
      "args": ["-y", "@fload/mcp"],
      "env": {
        "FLOAD_API_KEY": "your-api-key-here"
      }
    }
  }
}

Cursor — add the same configuration to your Cursor MCP settings.

That's it. The MCP server runs locally as a subprocess and connects to Fload using your API key.

What you can do

The MCP server gives your AI client access to 37 tools:

Apps

• List all apps in your organization
• Get detailed information about any app

Reviews

• Access reviews with filters for rating, platform, and date
• generate_review_reply — Generate an AI draft reply for a review
• send_review_reply — Send a reply to an app review
• translate_review — Translate a review to English

Analytics

• Get key metrics (revenue, downloads, subscriptions) for any date range

ASO

• get_aso_summary — Get ASO score, health, and overview for an app
• get_aso_recommendations — Get current ASO recommendations (title, subtitle, keywords)
• get_aso_keywords — Get keyword rankings, search volume, and competitor data
• get_aso_experiments — List ASO experiments with status filtering
• get_aso_locale_snapshots — Get current listing snapshots across all locales
• trigger_aso_analysis — Trigger a new ASO analysis run

Agents

• See which agents are configured and their status
• View recent agent run history and results
• trigger_agent_run — Trigger a manual agent run
• pause_agent — Pause an agent
• resume_agent — Resume a paused agent
• get_agent_activity — Get recent activity log for an agent

Growth

• Get the latest growth audit report and scores

Monitoring

• Access detected anomalies across all metrics
• get_anomaly_detail — Get full detail for a single anomaly
• acknowledge_anomaly — Mark an anomaly as acknowledged
• dismiss_anomaly — Dismiss an anomaly

Ads

• Get campaign performance data across all platforms

Forecasting

• Get revenue and download projections

Chat

• list_conversations — List chat conversations
• get_conversation_messages — Get messages in a conversation
• send_chat_message — Send a message to the AI chat

Dashboard

• High-level overview of all your apps at once

Pending actions

• List actions awaiting your approval
• Approve or reject actions directly through the AI

Example prompts

Once connected, just talk to your AI client naturally:

• "What happened to my revenue this week?" → pulls analytics data and compares to last week
• "Are there any anomalies I should know about?" → checks detected anomalies
• "What's my current growth score?" → retrieves the latest growth audit
• "Show me pending actions and approve the review replies" → lists and approves actions

API key management

• Keys are organization-scoped — they access all apps in your org
• Keys support configurable permission scopes (read-only, specific tools, or full access)
• Generate and revoke keys from Settings → API Keys
• See API Keys for full details

Limitations

• The server runs locally as a subprocess (STDIO transport)
• One organization per API key
• MCP reads and writes are free — they don't consume credits