ThinkingData MCP. Audit game events & user cohorts by conversation.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
ThinkingData / 数数科技 connects your AI client directly to a massive gaming analytics database. You audit user behavior, query specific event streams, and check profile data by just asking questions.
It lets you run complex reports—like tracking player funnels or auditing cohorts—without ever logging into the TA Dashboard.
What your AI agents can do
Get event properties
Retrieves the specific data properties associated with a defined project event.
Get project summary
Fetches an overview of the entire analytics project, including active events and cohorts.
List defined events
Lists all event names that have been defined within your specific data project.
Run queries on user IDs to get details like total level or last login time.
Execute detailed event queries to track specific player actions and conversion paths.
List all project events and retrieve their associated property definitions.
View existing user cohorts or update properties for specific users.
Manually track custom events and set user properties to test new features rapidly.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
ThinkingData / 数数科技: 8 Tools for Gaming Data Analysis
Use these eight tools to query project metadata, audit user profiles, map complex event sequences, and track custom data points directly through your agent.
019d848dget event properties
Retrieves the specific data properties associated with a defined project event.
019d848dget project summary
Fetches an overview of the entire analytics project, including active events and cohorts.
019d848dlist defined events
Lists all event names that have been defined within your specific data project.
019d848dlist project cohorts
Shows a list of user groups (cohorts) that have been predefined for segmentation.
019d848dquery events
Runs a complex query against event data to analyze player behavior and conversion funnels.
019d848dquery users
Executes a specific query on user profile data using distinct IDs.
019d848dset user properties
Updates or changes attributes in a user's profile record for targeted testing.
019d848dtrack custom event
Sends and logs a single, temporary custom event directly through the agent for immediate testing.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with ThinkingData / 数数科技, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
What you can do with this MCP connector
ThinkingData connects your AI client straight to a massive gaming analytics database. You audit user actions, check event streams, and pull profile data simply by asking questions. It lets you run complex reports—like tracking player funnels or auditing cohorts—without ever logging into the TA Dashboard.
Discovering Data Schema & Project Overview
The get_project_summary tool gives you a high-level snapshot of the entire analytics project, showing what events are active and what user groups (cohorts) exist. To see exactly which event names are defined in your data project, use list_defined_events. When you pick an event, the get_event_properties tool retrieves the specific data properties attached to it.
Mapping Player Funnels & Behavior Analysis
The query_events tool runs complex queries against all event data, letting you map out how players move through a game or app. You can check existing user segments by listing defined groups with list_project_cohorts.
Auditing and Managing User Profiles
You run specific checks on user profile records using distinct IDs via the query_users tool. To test new features, you change a user's attributes directly in their profile record by calling set_user_properties. If you need to simulate an action for testing, you can send and log a single, temporary custom event immediately using track_custom_event.
How ThinkingData MCP Works
- 1 Subscribe to the ThinkingData server and provide your TA Project APPID, Data Token, Query Secret, and API URL.
- 2 Instruct your AI client to run a data task (e.g., 'List all events') using one of the available tools.
- 3 The agent executes the query against the analytics backend and returns structured results directly in the chat.
The bottom line is: you tell your AI what data you need, and it runs the complex queries for you.
Who Is ThinkingData MCP For?
Anyone who has to dig through game logs or user metrics without a dedicated terminal session needs this. Think of the Live-Ops Manager scrambling at 2 AM to verify an event tracker, or the Data Analyst sick of writing repetitive SQL queries just to get a summary. This is for people whose job involves answering 'how?' about player behavior.
Runs query_events and list_defined_events to build reports on user retention and funnel drop-off rates.
Uses the agent to monitor real-time player behavior or update properties via set_user_properties during live events.
Integrates analytics checks into daily routines, validating event tracking using track_custom_event before deployment.
What Changes When You Connect
- Stop jumping between dashboards. You can run complex behavioral queries using
query_eventsand get the results immediately in your chat window, keeping context intact. - Know exactly what data you're working with. Use
list_defined_eventsto see every event name available, then useget_event_propertiesto check its schema before writing a query. - Manage user segments on the fly. You can list existing groups with
list_project_cohorts, and if you find a bug, correct it by runningset_user_propertiesfor a specific ID. - Test new features instantly. Instead of waiting for an engineer to run a script, use
track_custom_eventto simulate real user actions right from your prompt. - Get a bird's-eye view immediately.
get_project_summarygives you the project status—how many events, how many cohorts—in one single command.
Real-World Use Cases
Troubleshooting a Funnel Leak
A new feature is performing poorly. Instead of manually querying logs, ask your agent to run query_events for the 'PurchaseStart' event. The agent filters out all failed attempts and shows you which specific user actions precede the drop-off, pinpointing the exact behavioral bottleneck.
Auditing a VIP User
A key player reports an issue. You ask your agent to run query_users for their ID. The agent returns their profile data—level, cohort status, last login—allowing you to check if the bug is client-side or related to their stored metadata.
Validating Event Tracking
You just deployed a new mini-game. Before rolling it out, use track_custom_event to manually fire the event from your prompt. You confirm that the data hits the backend correctly and get immediate confirmation on schema integrity.
Preparing for Live Events
A live-ops update is coming, requiring a specific user group. First, use list_project_cohorts to confirm the group exists. Then, you query that cohort using query_events to get metrics on their current activity level.
The Tradeoffs
Running massive queries without knowing schemas
Just typing 'show me all user data' and hoping the database doesn't crash or returns useless columns. This wastes time and risks hitting rate limits.
→
First, run list_defined_events to see what exists. Then, use get_event_properties on a specific event (e.g., 'LevelUp') to verify the exact schema before running your full query via query_events.
Manually updating user data records
Having to log into a separate admin panel just to change one user's status or add a test property. This is slow and prone to human error.
→
Use the agent to run set_user_properties. You just need to tell it: 'Update USER_123 with {status: beta_tester}'—it executes the update command directly.
Assuming all events are tracked
Thinking a crucial event (like checkout failure) is being logged, but realizing later it wasn't defined or tracked correctly. The data gap goes unnoticed.
→
Use list_defined_events to create an inventory of what is tracked. If the required event isn't there, use track_custom_event for testing, and then escalate using that knowledge.
When It Fits, When It Doesn't
You should use this server if your core problem is 'I need to answer a complex question about player behavior or user data without manually clicking through multiple dashboards.' This tool excels at deep querying (using query_events) and profile auditing (query_users).
Don't use it if you only need simple, immediate system notifications or basic logging. For those tasks, a dedicated real-time monitoring dashboard or stream processor is better. If your goal is just to check API connectivity status, the project summary (get_project_summary) can give a quick look, but for ongoing operational checks, focus on the event data.
In short: Use this when you need deep pattern matching and historical context. Don't use it when you only need simple metrics or instant alerting.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by ThinkingData / 数数科技. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 8 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Sifting through raw game logs is a nightmare.
Today, figuring out why players are dropping off takes hours. You jump from the main dashboard to the event log, then you filter by user ID, then you try to match timestamps across three different tabs. It's copy-paste hell, and every time you click a new dropdown, you lose context of what you were looking at.
With this MCP server, you tell your agent exactly what you need—like 'Show me all users who viewed the tutorial but never bought anything.' The agent runs `query_events` in the background. You get a clean list and the metrics instantly. No clicks, no dashboard hopping.
ThinkingData MCP Server: Querying data using natural language.
Manual data retrieval requires specialized SQL knowledge or deep familiarity with the platform's specific UI structure. You have to know if you should use `query_events` vs. querying user profiles directly via `query_users`, and how those tools interact. It’s a steep learning curve.
This server handles that complexity for you. By exposing clear, actionable tools like `get_project_summary` and allowing direct query execution, your agent abstracts the technical difficulty away. You just ask the question; it finds the answer.
Common Questions About ThinkingData MCP
How do I list all available events using ThinkingData / 数数科技 MCP Server? +
Run list_defined_events. This tool fetches a comprehensive list of every event name your project is tracking, giving you an inventory before you build any queries.
What's the difference between query_users and query_events? +
query_users focuses on individual user records—things like total level or last login. query_events looks at actions taken by users over time, letting you map out funnels across many people.
Can I test a new event without deploying code? ThinkingData / 数数科技 MCP Server? +
Yes. Use track_custom_event. This allows your agent to simulate an event firing, letting you confirm the data structure and logging process instantly for testing purposes.
How do I find out what properties are available for a specific event? +
Use get_event_properties. You give it the event name, and it returns the schema, telling you exactly which fields (like 'purchase_amount' or 'level') can be queried.
What credentials do I need to set up the connection for ThinkingData / 数数科技 MCP Server? +
You must provide your TA Project APPID, Data Token, Query Secret, and API URL. These four values are required inputs to authorize all data operations through the server.
What key metadata does the `get_project_summary` tool give me about my ThinkingData project? +
The get_project_summary tool provides a high-level overview of your entire project. You retrieve details like its current operational status, the total count of defined events, and how many active user cohorts exist.
When should I use the `set_user_properties` tool in ThinkingData / 数数科技 MCP Server? +
Use set_user_properties when you need to manually update a specific user's profile data. This allows your agent to change properties for distinct IDs without needing to track a new event.
How does the `list_project_cohorts` tool help me with audience segmentation? +
The list_project_cohorts tool shows you every defined user cohort within your project. This lets you audit and manage specific groups of users for targeted analysis.
How do I find my TA Project APPID and Tokens? +
Log in to your ThinkingAnalytics Console, go to [Project Management] -> [Project Info] to find your APPID and Data Token. The Query Secret is usually found in the [Account Management] or [API Management] section.
What is the 'distinctId'? +
The distinctId is the unique identifier for a user in your system (e.g., a device ID or an account ID). It is used to tie events and properties to a specific individual profile in ThinkingAnalytics.
Can I perform complex data analysis through the agent? +
Yes! Use the query_events tool with a JSON query definition. Your agent will fetch the raw or aggregated results from TA, allowing you to audit behavioral patterns using natural language reasoning over the returned data.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Natural Tokenizer Engine
Tokenize text into words, numbers, emails, URLs, emojis, and hashtags deterministically. AI struggles with mixed content — this engine extracts exact linguistic entities instantly.
SimpleHash (Multi-chain NFT API)
Access comprehensive NFT data across 50+ chains — query wallet holdings, metadata, collection details, and transfer history.
UK ONS Population — Deaths, Well-being & Demographics
UK population and health statistics: weekly death registrations by age, sex, and region, personal well-being estimates (life satisfaction, happiness, anxiety), suicide data by local authority, and population projections from the ONS.
You might also like
DingConnect
Equip your AI agent to manage mobile top-ups, track operators, and monitor account balance via the DingConnect API.
AfterShip
Track packages worldwide — create, manage, and monitor shipments via AI.
Alexa Smart Home
Control Alexa-connected smart home devices — lights, thermostats, speakers, and sensors via Alexa Smart Properties API.