mcp-dashboards

Overview

MCP Dashboards is a comprehensive MCP server that transforms data into interactive charts, dashboards, and KPI widgets directly within AI conversations. It provides 31 tools covering 44+ chart subtypes with 21 built-in themes, enabling users to create everything from simple bar charts to complex dashboards without leaving their AI environment. The server supports live polling from APIs, exports to various formats (PPT, A4, PNG), and works across multiple AI clients including Claude Desktop, VS Code, and others.

Try asking AI

After installing, here are 6 things you can ask your AI assistant:

• render_pie_chart

  Creates pie/donut charts for composition data

  Try:Show my portfolio allocation as a donut chart
• render_bar_chart

  Creates bar charts for comparison data

  Try:Compare Q1 vs Q2 revenue by region as a bar chart
• render_line_chart

  Creates line/area charts for trends

  Try:Show sales trends over the last 12 months
• render_dashboard

  Creates multi-chart dashboards with responsive grid

  Try:Build a dashboard with monthly sales KPIs and a trend line
• render_geo_chart

  Creates geographic/chloropleth maps

  Try:Visualize website traffic by country on a world map
• render_live_chart

  Creates real-time charts that auto-update from APIs

  Try:Track Bitcoin price live, updating every 30 seconds
• render_funnel_chart

  Creates conversion funnel charts

  Try:Show the conversion funnel from signup to purchase
• render_radar_chart

  Creates radar charts for multi-axis comparison

  Try:Rank the team by performance using a radar chart
• render_from_json

  Auto-detects best chart for JSON data

  Try:Visualize this sales data: {JSON data}
• poll_http

  Fetches JSON from HTTP endpoints for data visualization

  Try:Get weather data from API endpoint
• render_from_url

  Fetches and visualizes data directly from a URL

  Try:Visualize the CSV data from this URL
• render_hero_metric

  Creates KPI widgets in 11 different styles

  Try:Show monthly revenue as a progress ring widget

Comparable tools

Installation

{
  "mcpServers": {
    "dashboard": {
      "command": "npx",
      "args": ["-y", "mcp-dashboards", "--stdio"]
    }
  }
}

claude mcp add dashboard -- npx -y mcp-dashboards --stdio

npx mcp-dashboards
# Server starts on http://localhost:3001/mcp

FAQ

Which AI clients support MCP Dashboards?

It works in Claude Desktop, Claude Web, VS Code (GitHub Copilot), Goose, Postman, and MCPJam. ChatGPT support is rolling out.

Can I use custom themes in my visualizations?

Yes, there are 21 built-in themes with options to customize palette, typography (8 options), and effects (5 presets).

Is there a way to access visualizations outside the AI client?

Yes, when the server detects your client can't render inline, it provides a localhost URL and a standalone HTML file for offline access.

Compare mcp-dashboards with