GameAnalytics 发布 MCP Server 支持 AI 工具直连游戏数据

What if your game data could answer you directly? For years, game teams have relied on dashboards, SQL queries, and scattered tools to understand what their players are doing. Useful? Absolutely. Fast and intuitive? Not always. As AI reshapes how developers work, we asked ourselves a simple question: what if your game data could just talk back?

Today, the GameAnalytics MCP Server is the answer. It’s a hosted bridge between your GameAnalytics data and AI assistants such as Claude, ChatGPT, Gemini, Cursor, Codex, and any other tool that supports the MCP standard. You add one URL (https://mcp.gameanalytics.com) to your AI tool, sign in with your GameAnalytics account through a standard OAuth flow, and start asking questions in plain language.

What the MCP server does

MCP is a standard that lets AI tools talk to external APIs in a structured way. The GameAnalytics MCP Server acts as the translator between your AI assistant and the GameAnalytics Metrics API. Instead of manually filtering dates and dimensions, you can open your preferred AI client and ask something like:

“Show me D7 retention for our iOS build last week, and compare it to the Android build.”

The server translates your request into the right API calls, pulls the data, and returns a summary your AI can reason about. Access is read-only. The server can see every Organization, Studio, and Game your account has access to, but it cannot change your data or settings.

Secure sign-in through OAuth

The first time you connect, your AI tool opens a sign-in page in your browser. You sign in with your GameAnalytics account, review a consent screen that names the AI tool asking for access, and choose Allow. From then on, the connection is live. You can revoke it any time by removing the server in your AI tool.

This is a significant change from how most game data integrations work today. There are no API keys to copy, no credentials to manage, and no custom auth setup. Standard OAuth, same pattern as signing in with Google.

Available two ways

The MCP Server is available two ways. Teams who want their game data inside their AI tools without the full data pipeline can get MCP Access on its own at $149 per month. PipelineIQ Pro includes MCP access alongside the Data Warehouse, Data Export, Metrics API, Organization API, and custom event fields.

What studios are doing with it

The MCP Server has already landed in production workflows across very different studios, and the ways teams are using it tell the story better than any example we could invent.

Blitz of Battle is using the server to go beyond standard reporting. The publishing team is exploring correlations between metrics that would be too tedious to surface manually, which opens up a layer of analysis that most reporting workflows skip entirely.

GameAnalytics MCP Server worked better than expected. I was able to get further analyses done by looking at correlations of metrics that I wouldn’t normally be looking at. - Publishing Director, Blitz of Battle

Pitch Labs is building autonomous agent pipelines around their product analytics. For teams building AI-native workflows, MCP has become infrastructure rather than a feature, and the Pitch Labs team moved on it the moment it launched.

MCP access is crucial in building any autonomous agent pipeline, especially for product analytics. The moment GameAnalytics launched the MCP server, we immediately upgraded our pipeline. - Derrick Ko, CEO and Co-Founder, Pitch Labs

Zero One plugs the server into an entire agent chain that runs inside Cursor. The chain handles everything from ideation to monetization, with the GameAnalytics MCP Server as the plug-and-play data layer that validates whether each release actually moved the needle.

The GameAnalytics MCP Server slots directly into our AI factory in Cursor, a chain of agents managing everything from ideas to monetization. A plug-and-play data source that validates whether each release actually moved the needle. - Quentin Sallat, Co-Founder, Zero One

Gamebit is using the server for recurring product work. The team runs weekly progress reports, compares builds against each other, and uses the server to surface what to prioritize in the next release, all through the same connection.

The GameAnalytics MCP Server reads data well and gives valuable advice for the project. We can build weekly progress reports, record our data, compare builds, and identify potential improvements in the codebase. It also advises on what to build in the next update, shows what we should focus on, and compares weak points across builds. This is a huge feature. - Huseyin Gulgen, Co-Founder, Gamebit

Why this matters

Every team member engages with data differently. Product managers want quick clarity on KPIs like retention and ARPU. Developers want to validate custom events directly from their IDE. Analysts want to iterate on ideas before building dashboards. And now, teams building autonomous pipelines want a reliable, authenticated data source their agents can call. The MCP Server supports all of them because it brings game data to where you already work, rather than asking you to come to the data.

Getting started

Setup takes a few minutes. Get MCP Access for your Organization, add the server URL to your AI tool, sign in through OAuth, and start asking questions. Full connection instructions for Claude, Claude Code, Cursor, Codex, and more are in the documentation.

If you want the MCP Server alongside full event-level data, data warehouse access, and raw data export, PipelineIQ Pro remains the right path.

What’s next

Conversational analytics is still early. The MCP Server is the connective tissue, and the use cases studios are building on top of it are evolving fast. Expect more capabilities, more tools, and more ways to integrate your game data into the AI workflows your team actually uses day to day.

‍