Steam Trends MCP

What Steam Trends MCP MCP server does

Steam Trends MCP MCP server gives AI agents access to Steam trend data through the TrendsMCP service. Its three tools cover different analysis patterns:

• get_time_series returns weekly history, with daily data available through the REST interface, for one source and keyword.
• get_growth calculates point-to-point percentage change for one keyword and source. Multiple comparison windows can be included in one call.
• get_top_trends returns a live ranked feed and does not use a keyword.

The historical series uses a 0–100 index where the underlying pipeline supports it; in that scale, 100 represents the peak within the returned period. Absolute volume values are included when the source provides them.

How it works

The hosted MCP endpoint is https://api.trendsmcp.ai/mcp. Requests use a Bearer API key, and the same key can be used with the REST endpoint at https://api.trendsmcp.ai/api. The repository also provides a stdio adapter for local MCP hosts and Glama deployments.

Each historical request takes one source and one keyword. get_growth accepts a single source or a comma-separated source list, with periods ranging from short windows such as 7D to longer options such as 5Y, plus calendar-oriented periods including MTD, QTD, and YTD. For live rankings, MCP calls require a type value matching the feed name exactly, including capitalization. Results can be limited, paginated, sorted by rank or rank change, and filtered by category where supported.

Setup and configuration

Create an account and obtain an API key before making paid requests. The hosted service can be added to clients that support remote MCP connections. For local stdio use, install the repository package and run python -m trends_mcp_server; configure TRENDSMCP_API_KEY in the MCP server environment.

Cursor, VS Code, Windsurf, Cline, and Claude Code can connect to the hosted URL with an authorization header. Claude Desktop does not provide native HTTP support in the documented setup, so the README uses mcp-remote as a bridge. Claude.ai’s custom connector uses the hosted URL with OAuth rather than placing a Bearer key in its connector configuration.

Tools and capabilities

Steam Trends MCP MCP server is suited to questions such as whether a game’s interest is rising, how its indexed activity changed over a selected period, or which items currently appear in a live ranking. get_time_series supports weekly history through MCP and weekly or daily modes through REST. get_growth defaults to a 12-month comparison when no period is supplied. get_top_trends defaults to 25 results and supports up to 200, with pagination through offset.

Limitations and notes

The service requires one source per get_time_series call. Keyword formats depend on the selected source. The type argument is mandatory for MCP get_top_trends requests and must use the feed’s exact name. On REST, leaving out type requests every feed and is billed per feed.

The README states a free allowance of 100 requests per month. The local stdio process can list tools without a key, but calls that access paid data require TRENDSMCP_API_KEY and use the same quota. Current parameter details are maintained in the linked TrendsMCP documentation.