Amplitude MCP for AI. Analyze product behavior and manage cohorts instantly.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Connect to your AI in seconds.
Amplitude MCP connects your AI agent directly to advanced product analytics. Send event data, fetch real-time user profiles, and query deep segmentation metrics—all without leaving your chat window.
Use Amplitude's full suite of tools to track how users interact with your product and pinpoint where they drop off.
What your AI can do
Create annotation
Marks specific dates directly onto product charts for better context.
Create category
Sets up a new group or type for your event data (Taxonomy).
Create release
Creates an official record to track feature changes and their impact on usage.
Send single or large batches of events to record specific user activity in the product.
Fetch a user's current profile and all associated properties in real time.
Identify specific groups of users based on shared actions or membership lists, managing behavioral cohorts.
Run complex queries against your data to find out how various user segments are performing.
Organize and maintain event categories, properties, and release records to keep the data clean and structured.
Ask an AI about this
Waiting for input…
Amplitude Alternatives: 21 Tools for Product Analytics
These tools let you manage the entire analytics lifecycle—from sending raw events to defining cohorts and exporting massive datasets.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Amplitude on VinkiusCreate Annotation
Marks specific dates directly onto product charts for better context.
Create Category
Sets up a new group or type for your event data (Taxonomy).
Create Release
Creates an official record to track feature changes and their impact on usage.
Delete Users
Submits formal data deletion requests required for compliance (GDPR/CCPA).
Export Data
Pulls all raw event records into a JSON file for deep, external analysis.
Get Audit Logs
Retrieves logs that track administrative activity within the analytics platform.
Get Experiment Variant
Checks which version of an A/B test a specific user was assigned to.
Get Segmentation
Queries and retrieves the metrics that are usually shown in Amplitude's chart...
Get User Profile
Fetches a user's complete, up-to-date profile data and associated properties.
Group Identify
Updates or sets descriptive properties for defined groups of users.
Identify User
Adds or changes a user's details without creating a new event record.
List Categories
Shows all existing event categories used in the product (Taxonomy).
List Cohorts
Retrieves a list of every behavioral group or cohort that has been defined.
List Event Properties
Lists all possible descriptive fields (properties) attached to events (Taxonomy).
List Event Types
Shows every type of event that can be tracked in the product (Taxonomy).
List User Properties
Retrieves all possible properties you can attach to a user's profile (Taxonomy).
Map User
Links one user ID to another, helping consolidate identity data.
Send Attribution
Sends marketing attribution data collected from partner sources into the platform.
Send Batch Events
Transmits a large volume of event records to the analytics cloud efficiently.
Send Events
Sends smaller, structured arrays of event objects directly to the platform.
Upload Cohort
Creates a new behavioral cohort using an uploaded list of user IDs.
Security and governance baked right in.
Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.
Claude AI
Open Claude Settings
Go to claude.ai, click your profile icon, then navigate to Customize → Connectors.
Add Custom Connector
Click the "+" button and select Add custom connector. Paste your Vinkius endpoint URL:
https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp
Replace [YOUR_TOKEN_HERE] with your token
from cloud.vinkius.com. For OAuth-protected servers, expand
Advanced settings to add credentials.
Start a conversation
Open a new chat. The Amplitude integration is available immediately — no restart needed.
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 Amplitude, then connect any of our 5,100+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,100+ 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
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Amplitude. 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 connection provides 21 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
The pain of manual metric verification
Right. So, you launch a new feature. You think it’s going to move adoption numbers, but figuring out if it actually worked is a nightmare. You have to jump between the dashboard, download raw data into a spreadsheet, and manually cross-reference user IDs against cohort lists just to prove your point. It's tedious clicking through tabs and copy-pasting metrics until you get an answer.
With this MCP, the entire process shifts. Instead of manual cleanup, you ask your agent directly: 'Show me how many users who were in the Beta Tester group adopted Feature Y.' The data comes back formatted, ready for review, cutting hours off what used to be a multi-day chore.
Getting real user insights with `get_user_profile`
Manual verification usually means opening the platform and searching for an ID, then checking every single property one by one. You're wasting time clicking through dozens of dropdowns just to piece together a complete picture of who that user is.
Now, you simply ask your agent to use `get_user_profile`. It pulls everything—all the properties and activity details—and hands it back in one clean package. That’s what's different now: instant, comprehensive visibility.
What your AI can actually do with this
This MCP lets you treat your product analytics platform like a conversation partner. You send natural language queries to get answers about user behavior, feature adoption, or retention rates. Instead of building complex dashboards or writing SQL queries just to check a metric, you talk to the data. Your agent handles the connection and pulls the specific insights you need—whether that's finding out which users are part of a high-value cohort or analyzing product performance after a new release.
It’s about getting answers instantly. You can analyze segmentation metrics or get user profiles on demand, turning raw event data into actionable knowledge right in your chat. If you're managing many different data sources, connecting through Vinkius makes it easy to access Amplitude alongside other services, keeping all your insights organized and accessible from one place.
019e5cfa-e7de-7101-9fcd-b19c4b3ebc06 
Here's how it actually works
The bottom line is that it brings complex, dashboard-driven analytics into your natural chat flow.
Subscribe to this MCP and provide your Amplitude API Key and Secret Key credentials.
Tell your AI agent exactly what you need—for example, 'Get the profile for user X' or 'Export raw events from last week.'
The agent executes the required tool calls using the stored keys and returns the specific data points in a readable format.
Who is this actually for?
Product Managers who hate leaving the chat to check adoption rates. Data Analysts tired of writing boilerplate SQL for basic cohort checks. Growth Engineers needing to validate A/B test variants on the fly.
Checking feature adoption or viewing event counts without having to navigate multiple dashboards.
Exporting raw data for a specific time frame or checking if the current event taxonomy is correct.
Identifying users who should be part of a new cohort, or checking which experiment variant they were assigned to.
What Changes When You Connect
You don't have to leave your chat to check metrics. Need a user profile? Use get_user_profile and get real-time data without opening a browser tab.
Manage your entire data structure from the terminal. Run list_event_types or list_user_properties to verify taxonomy before launching a feature, keeping your data clean.
Handle high volume data easily. Use send_batch_events for bulk uploads instead of running multiple small API calls—it’s faster and cleaner.
Compliance is simple. When you need to delete user records due to GDPR, just invoke delete_users. It's a one-step process.
Build out complex groups fast. Instead of manual list creation, run upload_cohort with a static ID list to instantly define a new segment.
See it in action
Diagnosing low feature adoption
A PM notices the 'checkout' funnel is dropping off. They ask their agent, 'What are the segmentation metrics for users who hit Step 3 but never converted?' The MCP runs get_segmentation and identifies that users who failed to load recommendations were the biggest drop-off group.
Auditing user data for compliance
A Growth Engineer needs to prove GDPR compliance. They use list_user_properties to confirm all PII fields are tracked and then run delete_users, initiating the secure deletion process right from their agent.
Analyzing a new campaign's impact
A data analyst launches a new marketing push. They ask the agent to 'Export raw event data for all users who clicked the coupon code.' The MCP executes export_data, delivering a JSON file containing only that specific activity.
Validating experiment success
A PM needs to check if an A/B test is working. They ask, 'What variant was user 456 assigned?' The MCP calls get_experiment_variant and confirms the user landed in the control group, allowing them to pivot their strategy.
The honest tradeoffs
Trying to get everything at once
Writing a massive prompt like 'Give me all data on users who signed up last month and are premium and clicked X.' This often fails or hits rate limits because the request is too broad.
Break it down into steps. First, check cohorts using list_cohorts to find the group ID. Then, fetch user details with get_user_profile, and finally, query metrics with get_segmentation. Keep the calls targeted.
Manually updating user properties
Updating a user's status (like 'VIP') requires going into the platform GUI, finding the ID, and clicking multiple times.
Use identify_user to update that property directly via your agent. You just tell it: 'Update user 123's property to VIP.' It handles the backend call.
Forgetting data structure rules
Trying to analyze a metric (like conversion rate) when the underlying event taxonomy hasn't been updated for the new feature.
Before querying, always check the available fields using list_event_properties and ensure you have registered any necessary changes by running create_category or create_release first.
When It Fits, When It Doesn't
Use this MCP if your job requires translating vague business questions ('Why are users leaving?') into specific, verifiable data requests (e.g., 'Show me the count of drop-offs after event X'). You need it when you must analyze user behavior and cohort membership without opening a dedicated analytics dashboard.
Don't use this if all you need is to query simple relational database records that don't involve behavioral tracking, or if you are only dealing with transactional data (like order totals). For pure ETL tasks, a direct database connector is better. If the goal is simply storing events in bulk and not analyzing them, then basic event streaming tools suffice. This MCP handles the analysis layer over the stored data.
Questions you might have
How do I export raw data using `export_data`? +
You specify the exact time range and event types you want to include in your request. The MCP then compiles all relevant records into a downloadable JSON file for deep analysis.
What is the difference between `send_events` and `send_batch_events`? +
send_events handles smaller, structured groups of events. Use send_batch_events when you have a massive volume of data to push into the cloud quickly.
Can I check if a user was in an A/B test using `get_experiment_variant`? +
Yes. You provide the unique ID, and the MCP returns which specific variant (A, B, or Control) that user was assigned to by the testing system.
How do I find out what properties are available for users? Use `list_user_properties`. +
Running list_user_properties gives you a full list of every custom field or property type that can be attached to any user profile in the system.
How do I handle data deletion requests for a user using `delete_users`? +
The delete_users tool submits formal data removal requests, ensuring compliance with regulations like GDPR or CCPA. It initiates the necessary process to scrub personal identifying information from your account.
If I need to update a user's profile without generating an event, should I use `identify_user`? +
Yes, that's exactly right. Using identify_user updates properties directly on the user record. This lets you correct metadata or set state information without triggering any event volume in your data stream.
How can I see all available top-level event categories using `list_categories`? +
The list_categories tool fetches the complete taxonomy of your system. This is crucial for understanding the structure and ensuring that new events you send fit into an existing, approved category.
What kind of activities are recorded when I use `get_audit_logs`? +
The get_audit_logs tool pulls records of administrative actions taken within your account. This lets you track who changed what, and when, providing a full audit trail for governance checks.
Can I export raw event data for a specific time range? +
Yes. Use the export_data tool by providing the start and end times in YYYYMMDDTHH format to retrieve raw event data in JSON format.
How do I check which cohorts a specific user belongs to? +
You can use the get_user_profile tool with the specific Amplitude User ID. It will return the user's profile, recommendations, and all active cohort memberships.
Is it possible to update user properties without sending a new event? +
Yes, the identify_user tool allows you to set, unset, or append user properties via the Identify API without triggering a separate event.
We've already built the connector for Amplitude. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 21 tools are live and waiting.
You're up and running in seconds.
Gemini Vinkius gives your AI agents access to the full catalog of app connectors, all fully managed, secure, and enterprise-ready. One subscription, every tool you need.
Built, hosted, and secured by Vinkius. You just connect and go.