MCP Guide

SteamData Connect supports the Model Context Protocol (MCP), enabling AI applications to directly access Steam game data.

What is MCP?

Model Context Protocol (MCP) is an open protocol that allows AI applications (such as Claude Desktop, Cline, etc.) to access external data and tools through a standardized interface. With MCP, AI can directly query Steam game information, review data, time-series statistics, and more, without manual copy-pasting.

Prerequisites

A SteamData account
Created team with generated API key (see Quick Start)
MCP-compatible AI client (such as Claude Desktop, Claude Code, GitHub Copilot, Cline, etc.)

Configuring MCP Server

MCP Endpoint Information

URL: https://endpoint.steamdata.ai/mcp
Authentication: Bearer Token
API Key Format: Authorization: Bearer sdk_your_key_here

Claude Desktop Configuration

Add to Claude Desktop's configuration file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "steamdata": {
      "url": "https://endpoint.steamdata.ai/mcp",
      "headers": {
        "Authorization": "Bearer sdk_your_api_key_here"
      }
    }
  }
}

Cline (VS Code Extension) Configuration

Add MCP server in VS Code's Cline settings:

{
  "mcp.servers": [
    {
      "name": "SteamData",
      "url": "https://endpoint.steamdata.ai/mcp",
      "headers": {
        "Authorization": "Bearer sdk_your_api_key_here"
      }
    }
  ]
}

Claude Code (VS Code Extension) Configuration

Use Claude Code CLI to add the MCP server:

claude mcp add --transport http steamdata https://endpoint.steamdata.ai/mcp \
  --header "Authorization: Bearer sdk_your_api_key_here"

Managing servers:

# List all configured servers
claude mcp list

# Get details for a specific server
claude mcp get steamdata

# Remove a server
claude mcp remove steamdata

For detailed configuration instructions, see the Claude Code Official Documentation.

GitHub Copilot Configuration

Configure in VS Code or other editors that support GitHub Copilot:

Settings Path: Settings → Extensions → GitHub Copilot → MCP Integration

{
  "github.copilot.advanced": {
    "mcpServers": {
      "steamdata": {
        "url": "https://endpoint.steamdata.ai/mcp",
        "headers": {
          "Authorization": "Bearer sdk_your_api_key_here"
        }
      }
    }
  }
}

Language Parameter Specification

All MCP tools with language parameters support multiple formats (standard names, BCP 47 codes, Steam API codes).

For detailed language parameter formats and complete mapping table, see Language Parameter Specification.

Available Tools

SteamData Connect provides the following MCP tools:

1. steam_get_game

Get detailed game information including basic info, pricing, review statistics, follower count, etc.

Parameters:

appid (required): Steam App ID

Example:

Please show me detailed information for Call of Duty: Black Ops 6 (App ID: 1938090)

2. steam_get_game_localized

Get localized game name and description.

Parameters:

appid (required): Steam App ID
language (optional): Language parameter, supports standard names, BCP 47 codes, or Steam API codes (see language mapping table above)

Example:

Get the Simplified Chinese name and description for Dota 2 (App ID: 570)

3. steam_batch_games

Batch fetch basic information for multiple games (up to 100).

Parameters:

appids (required): Array of App IDs

Example:

Compare basic information for these three games: 570, 730, 1938090

4. steam_search_games

Search games by name, supports multi-language search.

Parameters:

query (required): Search keyword
language (optional): Search language (see language mapping table above)
limit (optional): Result limit (1-100, default 20)

Example:

Search for all games with "Call of Duty" in their name

5. steam_filter_games

Get game list with flexible filtering, supports pagination and sorting.

Parameters:

filter: Filter criteria object
- app_type: Application type (game, dlc, demo, etc.)
- is_released: Whether released
- is_free: Whether free
- developer: Developer name
- publisher: Publisher name
- release_date: Release date range
- price: Price range
- review_score: Review score range
- And more filtering options...
page: Page number (starting from 0)
page_size: Items per page (1-100)

Example:

Find action games released in 2024 with review scores above 80, sorted by release date

6. steam_get_timeseries

Get historical time-series data for a game, including reviews, sales, revenue, player counts, etc.

Parameters:

appid (required): Steam App ID
start_date (required): Start date (YYYY-MM-DD)
end_date (required): End date (YYYY-MM-DD)
metrics (optional): Metrics list (review, sales, revenue, price, follower, wishlist, ccu)

Note: Sales and revenue data require special permissions.

Example:

Get review and player count data for Dota 2 in January 2024

7. steam_get_reviews

Get game user reviews list with flexible filtering and sorting.

Parameters:

appid (required): Steam App ID
language (optional): Review language (see language mapping table above)
voted_up (optional): Whether positive review (true/false)
steam_purchase (optional): Whether from verified purchasers
date_from (optional): Start date (YYYY-MM-DD)
date_to (optional): End date (YYYY-MM-DD)
order_by (optional): Sort field (timestamp_created, rid)
order_dir (optional): Sort direction (ASC, DESC)
offset (optional): Pagination offset
limit (optional): Result count (1-100, default 20)

Example:

Get the latest 50 English positive reviews for Black Myth: Wukong

8. steam_get_review

Get detailed information for a specific review by Review ID.

Parameters:

rid (required): Review ID

Example:

View details of review ID 12345678

Usage Examples

Example 1: Game Market Analysis

Prompt: Analyze the characteristics of the top 10 popular games released in 2024,
including their genres, price ranges, and user review trends

AI will automatically:

Use steam_filter_games to filter 2024 releases
Use steam_get_game to get detailed info for each game
Use steam_get_reviews to analyze user reviews
Generate comprehensive analysis report

Example 2: Competitive Comparison

Prompt: Compare CS:GO (730) and Valorant (1517290),
these two FPS games' player reviews and popularity trends

AI will automatically:

Use steam_batch_games to get basic info for both games
Use steam_get_reviews to analyze their user reviews
Use steam_get_timeseries to get historical data
Generate comparative analysis

Example 3: Localized Content Query

Prompt: View Elden Ring (1245620)'s Simplified Chinese introduction,
and analyze Chinese players' reviews

AI will automatically:

Use steam_get_game_localized to get Chinese content
Use steam_get_reviews to filter Chinese reviews
Analyze review trends and common feedback

Quota and Limits

MCP calls share quota with REST API:

Free Tier: 10,000 requests per month
Standard Tier: 100,000 requests per month
Enterprise Tier: Custom quota

Each MCP tool call counts as one API request. You can view real-time quota usage in Dashboard.

Error Handling

MCP protocol uses JSON-RPC 2.0 standard error format:

{
  "jsonrpc": "2.0",
  "error": {
    "code": -32001,
    "message": "Game not found",
    "data": {
      "appid": 999999
    }
  },
  "id": 1
}

Common error codes:

Error Code	Description
-32700	Parse error (invalid JSON)
-32600	Invalid request
-32601	Method not found
-32602	Invalid parameters
-32603	Internal error
-32001	Data unavailable (e.g., game not found)
-32002	Permission denied
-32003	Quota exhausted

Best Practices

Batch Queries: When querying multiple games, prefer steam_batch_games over multiple steam_get_game calls
Filter Optimization: Use precise filter criteria in steam_filter_games to reduce unnecessary data transfer
Cache Results: Game basic information (like name, publisher) changes infrequently, can be cached
Date Ranges: When querying time-series data, set reasonable date ranges to avoid requesting excessive data
Pagination: Use pagination parameters when handling large result sets to avoid request overload

Development & Debugging

Test MCP Connection

Test MCP endpoint using cURL:

curl -X POST https://endpoint.steamdata.ai/mcp \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sdk_your_key_here" \
  -d '{
    "jsonrpc": "2.0",
    "id": 1,
    "method": "tools/list",
    "params": {}
  }'

Test Tool Call

curl -X POST https://endpoint.steamdata.ai/mcp \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sdk_your_key_here" \
  -d '{
    "jsonrpc": "2.0",
    "id": 1,
    "method": "tools/call",
    "params": {
      "name": "steam_get_game",
      "arguments": {
        "appid": 570
      }
    }
  }'

Get Help

With MCP, you can easily build intelligent game data analysis applications! 🚀