# Kameleoon MCP for AI Agents AI Agent Connect > Kameleoon MCP lets you manage your A/B testing and personalization workflows through your AI agent. Instead of clicking through a complex dashboard to check experiment statuses or pull results, you can just ask your agent to do it. It connects your AI client directly to your Kameleoon account, making it easy to monitor live variations, check audience segments, and get performance reports without leaving your workspace. ## Overview - **Category:** developer-tools - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_6gJb8h8nvE83qpO01DFnHsxpfAmKyDZOllD6tBbY/ai-agent-connect - **Tags:** ab-testing, personalization, feature-flagging, experimentation, audience-segmentation, conversion-optimization ## Description Managing a high-volume experimentation roadmap usually means having dozens of tabs open. You're constantly jumping between the dashboard to check if a variation is actually live, looking up specific targeting rules for a new launch, or hunting for the latest results to share with the team. This Connector brings your experimentation data directly into your AI client. You can ask your agent to list all active experiments, pull specific site details, or check which audience segments are currently being targeted. If you need to know how a specific variant is performing, you can request a results report on the fly. It's built to give you a conversational interface for your growth stack. By connecting this through the Vinkius catalog, you turn your AI into a dedicated growth operations assistant that handles the data fetching so you can focus on making decisions. Instead of manually navigating complex menus to find a single piece of data, you can just describe what you need to the agent. It handles the heavy lifting of querying the platform, giving you the information you need to move faster. ## Tools ### create_experiment Create a new experiment by providing a name and the target site ID. This lets you launch new tests without leaving your workspace. ### list_custom_data View all custom data dimensions available in your account. This helps you see what specific metrics you can track for your tests. ### list_experiments View all experiments to monitor campaign statuses and find active tests. You can quickly see what's running across your entire account. ### get_site Fetch the metadata and details for a specific site. Use this to quickly grab the details of a specific environment. ### get_experiment_results Request a results report for an experiment to see how it's performing. This is the fastest way to get performance data for your reports. ### list_targeting_rules View the targeting rules used to allocate traffic to different variants. This helps you audit who is seeing which version of your site. ### list_segments List all audience segments defined in your account. Use this to see how your traffic is being split into groups. ### list_sites Get a full list of all sites registered in your Kameleoon profile. This is great for verifying your site inventory in one go. ### list_variations See the different variations associated with a specific experiment. Use this to check exactly what options are available for a test. ### get_experiment Pull the specific details and configuration for a single experiment. This gives you a deep dive into the settings of any test. ## Prompt Examples **Prompt:** ``` List all active experiments in my account. ``` **Response:** ``` I've fetched your experiments. Currently, you have 4 active experiments: * **New Checkout Flow (AB)** * **Home Page Personalization** * **Summer Sale Banner** * **Login Redesign** Which one would you like to inspect? ``` **Prompt:** ``` What are the variations for experiment ID '12345'? ``` **Response:** ``` For experiment 12345, I found 3 variations: | Variation Name | Status | | :--- | :--- | | **Reference (Original)** | Live | | **Variant A (Red Button)** | Live | | **Variant B (Blue Button)** | Live | All variants are currently receiving traffic. ``` **Prompt:** ``` Show me the targeting rules for our current audience segments. ``` **Response:** ``` Here are the active targeting rules for your segments: * **Segment A**: Users from US/CA * **Segment B**: Returning visitors * **Segment C**: Mobile-only users Let me know if you need the specific rules for any of these. ``` ## Capabilities ### Create new experiments Launch new tests by providing a name and the target site ID directly through your AI client. ### Pull performance reports Request the latest results for a specific experiment to see how your variants are performing. ### Check audience segments View all defined audience segments and targeting rules used to allocate your traffic. ### List experiment variations See all variations associated with a specific experiment ID to check your current options. ### View targeting rules Audit the rules used to target specific users across your different digital properties. ### Identify active experiments Monitor all active campaigns and identify live tests in your account in one view. ### Fetch site details Get metadata and configuration details for any site registered in your profile. ## Use Cases ### Quick Status Check A PM asks the agent to check if the new checkout variants are live. The agent uses `list_experiments` to confirm the status in seconds. ### Targeting Audit A growth engineer needs to know the rules for the 'Summer Sale' segment. The agent calls `list_targeting_rules` to show the exact traffic split. ### Automated Reporting An analyst asks for the latest results of the 'Home Page Test'. The agent triggers `get_experiment_results` to pull the latest data. ### Site Inventory Verification A developer needs to find the ID for the 'Marketing Blog' site. The agent uses `list_sites` to provide the correct ID instantly. ## Benefits - Skip the dashboard fatigue by using `list_experiments` to see what's live in seconds. You can check your entire roadmap with a single prompt instead of clicking through multiple menu levels. - Get instant answers on targeting with `list_targeting_rules` instead of digging through menus. This helps you verify who is seeing which version of your site without any manual searching. - Pull performance data quickly using `get_experiment_results` for your weekly updates. You can grab the latest numbers for any specific test and share them with your team instantly. - Verify your site inventory and metadata instantly with `list_sites` and `get_site`. This ensures you are always targeting the correct environment when launching new growth initiatives. - Keep your variations organized by quickly pulling data with `list_variations`. You can see all available options for a test ID to ensure your agent has the full context of the experiment. - Speed up your launch process by creating tests directly with `create_experiment`. This allows you to initiate new A/B tests and personalization workflows through natural language commands. ## How It Works The bottom line is you get a direct line to your experimentation data without the dashboard fatigue. 1. Subscribe to the Kameleoon MCP via the Vinkius catalog. 2. Enter your Kameleoon Client ID and Client Secret into your AI client settings. 3. Ask your agent to list experiments or pull results to start managing your growth tests. ## Frequently Asked Questions **Can I use the Kameleoon MCP to create new tests?** Yes, you can use the `create_experiment` tool to set up new tests via your AI client. This lets you launch new A/B tests using natural language commands. **How do I see if my A/B tests are actually live?** Your agent can use `list_experiments` to give you a real-time status of every campaign in your account. You'll see exactly what's running without opening the dashboard. **Can this Connector pull my conversion data?** Yes, you can request results reports using `get_experiment_results` to see how your variants are performing. It's the fastest way to get performance data for your reports. **Does this work for multiple sites?** Yes, it can query your entire site inventory using `list_sites` and `list_segments`. This makes it easy to manage experiments across different digital properties. **Is this for designing the variations?** No, this Connector is for managing, querying, and reporting on experiments, not for building the actual visual designs or creating the content of the variations. **How do I connect my account?** You just need your Kameleoon Client ID and Secret to link your account to your AI client. Once connected, your agent can access all the experimentation data you need. **How can I check if an experiment is currently active on a specific site?** Use the `list_experiments` tool to get a summary of all experiments, which includes their current status (e.g., 'ACTIVE', 'PAUSED') and assigned site IDs. **Can I see the defined variations for a particular experiment?** Yes, the `list_variations` tool takes an experiment ID and returns all variants configured for that test, along with their metadata. **Is it possible to list targeting rules and segments?** Absolutely. You can use the `list_segments` and `list_targeting_rules` tools to inspect how your traffic is being allocated across your experimentation ecosystem.