# PostHog MCP for AI Agents AI Agent Connect > PostHog lets you connect your product analytics to any AI agent so you can manage feature flags and track user behavior through natural conversation. Get real-time insights on retention, funnels, and experiments without switching tabs. ## Overview - **Category:** data-analytics - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_vpYk0NiKfoRYdh0mOuVMNfGBNKFlD4K0TVsVAy1i/ai-agent-connect - **Tags:** product-analytics, feature-flags, user-behavior, event-tracking, funnel-analysis, retention-metrics ## Description PostHog lets you connect your product analytics to any AI agent so you can manage feature flags and track user behavior through natural conversation. This Connector turns your data into a conversational resource. Instead of hunting through nested menus or manual filters, you can just ask your agent to tell you how a specific feature is performing or who your most active users are. It makes it easy to see if a toggle is live or to pull up a trend from a previous experiment without leaving your current workspace. Because Vinkius hosts this, you can get this connected and running in minutes. It's built for people who need to move fast and want the facts without the friction of traditional dashboard navigation. Don't waste time jumping between tabs just to verify a single metric. You can check if your tracking is working by looking at recent events or pull up a list of identified users to see their specific properties. It handles the heavy lifting of data retrieval so you can focus on making decisions based on what the numbers actually say. It's about getting the right answer at the right time. ## Tools ### get_event Use get_event to pull specific details for a single event. This is perfect for deep-diving into one user action. ### get_insight Use get_insight to fetch the full details of a specific insight. Use this to get the exact numbers behind a trend. ### get_person Use get_person to retrieve the full profile of a specific person. Use this to see all properties for a single user. ### list_actions Use list_actions to see the defined user actions in your project. This helps you understand how events are categorized. ### list_dashboards Use list_dashboards to see all the dashboards in your project. Use this to get an overview of your reporting structure. ### list_events Use list_events to see the most recent events from your project. This helps you verify that your tracking code is firing correctly. ### list_feature_flags Use list_feature_flags to see every feature flag in your project. This allows you to quickly see which toggles are enabled. ### list_insights Use list_insights to pull up your saved insights. This lets you grab data on funnels and retention quickly. ### list_persons Use list_persons to find identified users in your project. This is useful to look at specific user behavior patterns. ### list_experiments Use list_experiments to see all active and past experiments. This helps you track long-term product improvements. ## Prompt Examples **Prompt:** ``` List all saved insights in our PostHog project. ``` **Response:** ``` I've retrieved 12 saved insights. Some of the most recent ones are: * **Daily Active Users** * **Signup Funnel** * **Feature Adoption Trend** Would you like me to pull the specific numbers for any of these? ``` **Prompt:** ``` Check the status of all feature flags. ``` **Response:** ``` You have 8 feature flags in your project: | Flag Name | Status | | :--- | :--- | | new-onboarding-flow | **Active** | | beta-search | **Active** | | legacy-ui-toggle | Disabled | | dark-mode-test | Disabled | 5 are currently active and 3 are disabled. ``` **Prompt:** ``` List the last 5 persons identified in our project. ``` **Response:** ``` I've fetched the latest identified users. Here are the 5 most recent persons: 1. **User_8821** (Signed up: 2026-07-09) 2. **User_9930** (Signed up: 2026-07-09) 3. **User_1024** (Signed up: 2026-07-08) 4. **User_1155** (Signed up: 2026-07-08) 5. **User_1200** (Signed up: 2026-07-07) I can pull the full properties for any of these users if you need more detail. ``` ## Capabilities ### Query saved insights Get data on trends, funnels, and retention charts instantly. ### Track identified persons See specific user properties and behavior patterns. ### Manage feature flags Check the status and configuration of your toggles. ### Monitor experiments View active and past A/B tests to see what's working. ### Audit recent events Verify data ingestion and user actions from your logs. ## Use Cases ### Checking a live feature status A Product Manager asks their agent to check if the 'new-onboarding-flow' flag is live. The agent uses list_feature_flags to confirm the status. ### Verifying tracking for a new button A developer wants to see if a new click event is firing. They ask the agent to list recent events using list_events to verify the data. ### Analyzing a specific user profile A growth lead needs to see the properties of a specific user who reported a bug. They ask the agent to fetch the person using get_person. ### Reviewing experiment results A data analyst needs a summary of past tests. They ask the agent to list experiments to see all active and past results. ## Benefits - Stop manual dashboard filtering by using list_insights to get trend data instantly. - Verify tracking health immediately with list_events to see real-time data ingestion. - Check feature rollout status without leaving your chat using list_feature_flags. - Analyze specific user journeys by pulling person properties with list_persons. - Track experiment progress across your product using list_experiments to see what is working. - Audit your data integrity by pulling specific event details with get_event. ## How It Works The bottom line is you get a conversational interface for your entire PostHog analytics suite. 1. Subscribe to the PostHog MCP on Vinkius. 2. Enter your PostHog Personal API Key and Project ID. 3. Start asking your AI client questions about your product data. ## Frequently Asked Questions **How can I use PostHog with my AI agent for analytics?** You can connect your PostHog project to your AI client using this Connector. It allows your agent to query your data, check feature flags, and list user properties using natural language commands. **Can the PostHog MCP help me manage feature flags?** Yes, it can. You can ask your agent to list all your feature flags or check the status of a specific one to see if it is currently enabled or disabled. **Is it possible to see user behavior in PostHog using an AI agent?** Yes, the Connector allows your agent to list identified persons and retrieve their specific properties. This helps you understand individual user actions and behavior patterns. **Can I use this Connector to check if my tracking is working?** You can ask your agent to list the most recent events from your project. This is a quick way to verify that your tracking code is firing and capturing data correctly. **What kind of product insights can I get with PostHog and an AI agent?** You can retrieve saved insights regarding trends, funnels, and retention. Your agent can pull these up instantly so you don't have to build new reports every time. **Does the PostHog MCP support experiment monitoring?** Yes, it can list all active and past experiments. This makes it easy to review your A/B test results and product improvements through a simple chat interface. **Where do I find my Project ID?** You can find your Project ID in PostHog under **Project Settings**. It is a numeric ID usually found at the top of the settings page. **Does this support PostHog EU instances?** Yes! Use the optional `POSTHOG_HOST` credential and set it to `https://eu.posthog.com`. By default, it uses the US host (`https://app.posthog.com`). **Can I see individual user properties?** Absolutely. Use the `get_person` tool with a Person ID to retrieve all properties, tags, and identification metadata for a specific user.