# Unleash (Feature Toggles) MCP for AI Agents AI Agent Connect > Unleash (Feature Toggles) lets you manage feature flags, rollout strategies, and environment configurations directly through your AI agent. Instead of digging through a dashboard, you can ask your agent to check a flag's status, list all projects, or report usage metrics on the fly. ## Overview - **Category:** productivity - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_D9thr7rSK8ANvNy5GINLhbWdWBWkiOVYfkGJb9oC/ai-agent-connect - **Tags:** feature-flags, feature-management, deployment-strategies, rollout-control, environment-management, software-development ## Description Managing feature flags usually means jumping between different browser tabs, hunting for the right project, and squinting at rollout percentages. It's a tedious process that slows down deployment and creates room for human error. This Connector changes that by letting you handle those tasks through a simple conversation with your AI client. It pulls your data into your immediate workspace, making it easier to verify configurations before they hit production. You can even register new SDK instances and report usage metrics as you go. Because Vinkius hosts this Connector, you can plug it into your existing setup and start using it across your whole team's workflow. It turns a manual 'look it up' task into a 'just ask' action. You get to verify your rollout logic in real time, which means fewer broken builds and faster shipping cycles for your product. It helps you stay focused while keeping your feature management accurate. You can quickly confirm which features are live for specific users without the usual headache of manual searching. It removes the need to constantly toggle between your development environment and your management dashboard. ## Tools ### list_environments See every environment configured in your account. This helps you keep track of different deployment stages in one place. ### list_project_features Get a full list of all feature flags within a specific project. This is useful for auditing current capabilities. ### list_projects Pull a list of all projects you have active in your account. Use this to get a high-level overview of your work. ### list_segments View all defined segments for targeted feature rollouts. This helps you see who is currently in a test group. ### list_users Retrieve a list of all users configured in the system. Use this to verify specific user identities. ### get_client_features Fetch all feature flags and strategies for server-side evaluation. This allows for quick context checking. ### report_client_metrics Send flag usage data from your backend SDK to your management tool. This keeps your usage data accurate. ### register_client Create and register a new backend SDK instance. This is the first step for reporting backend metrics. ### get_frontend_features Get enabled flags for a specific context, like a user ID or property. This is perfect for real-time debugging. ### report_frontend_metrics Send usage data from your frontend SDK to your management tool. Use this to track frontend engagement. ### register_frontend Register a new frontend SDK instance in your account. This ensures your frontend data is properly tracked. ## Prompt Examples **Prompt:** ``` What flags are active in the production environment for the mobile app? ``` **Response:** ``` I've checked the production environment for the Mobile-App project. Here are the active flags: | Feature Flag | Status | Rollout Strategy | | :--- | :--- | :--- | | **biometric-login** | Enabled | 100% Rollout | | **push-notifications-v3** | Enabled | 50% Beta | | **offline-mode** | Disabled | No Rollout | Let me know if you need the specific targeting rules for any of these. ``` **Prompt:** ``` Is the 'dark-mode' feature enabled for user_99? ``` **Response:** ``` Evaluating flags for **user_99**... Yes, the `dark-mode` flag is currently **Enabled** for this user context. Other features like `experimental-chat` remain disabled. ``` **Prompt:** ``` List all the projects we have in the system. ``` **Response:** ``` I found 3 active projects: * **Default** (Main project) * **Mobile-App** (Flags for iOS/Android) * **Beta-Testing** (Experimental features) Would you like to see the flags for any of these? ``` ## Capabilities ### Check feature flag statuses Ask your agent to see which flags are currently active for specific user contexts. ### Audit project environments Get a full overview of your active environments and project structures. ### Fetch rollout strategies View the specific rules governing how features are rolled out to your audience. ### Report SDK usage metrics Send real-time usage data from your backend or frontend to your management tool. ### Manage user segments Verify which user groups are targeted for specific feature releases. ### Register frontend instances Add new frontend SDK instances to your account through a simple command. ## Use Cases ### Verifying specific project flags A developer needs to see every flag in a specific project to ensure a rollout is complete. They ask their agent to use list_project_features. ### Debugging user-specific features A QA engineer needs to know what a specific user sees on the live site. They ask the agent to use get_frontend_features for that user's ID. ### Auditing user access A security lead needs to verify which users are currently in the system. They ask the agent to use list_users. ### Quick environment checks A developer is unsure which project they are currently targeting. They ask the agent to list all projects to confirm. ## Benefits - Check your deployment status instantly by viewing all environments with list_environments. - Keep a high-level overview of your work by pulling all active projects with list_projects. - Verify targeted rollouts in seconds by pulling segments with list_segments. - Keep your usage data accurate by reporting backend data via report_client_metrics. - Speed up your setup by registering new backend instances with register_client. - Track your frontend engagement easily by reporting usage via report_frontend_metrics. ## How It Works The bottom line is you get instant access to your feature management data without leaving your primary workspace. 1. Enter your connection URL and token into the configuration. 2. Connect the Connector to your favorite AI client like Claude or Cursor. 3. Ask your agent to pull data or report metrics. ## Frequently Asked Questions **Can I check feature flags with the Unleash (Feature Toggles) MCP?** Yes. You can ask your agent to check the status of any flag, see which ones are enabled for specific users, or view the rollout strategies for your projects. **Does Unleash (Feature Toggles) MCP support frontend and backend?** It does. You can use it to register and report metrics for both your frontend and backend SDKs, ensuring you have a full picture of your feature usage. **How do I report metrics using the Unleash (Feature Toggles) MCP?** You can simply tell your agent to report usage data from your client. It will handle the communication with your account to keep your metrics up to date. **Can I see which segments are active in Unleash (Feature Toggles) MCP?** Yes, you can ask your agent to list all segments. This helps you verify which user groups are currently targeted for specific feature releases. **Does the Unleash (Feature Toggles) MCP work with Cursor?** Yes, it works with Cursor and any other MCP-compatible client like Claude or Windsurf. You can manage your flags directly from your code editor. **What happens when I register a client with Unleash (Feature Toggles) MCP?** Your agent will register a new SDK instance in your account. This is necessary for your app to start reporting metrics correctly. **Can I evaluate feature flags for a specific user ID?** Yes. Use the `get_frontend_features` tool and provide the `userId`. The agent will return the enabled flags based on the Unleash context for that specific user. **How do I see all feature flags associated with a specific project?** You can use the `list_project_features` tool by providing the `projectId`. This will list all toggles, their types, and current statuses within that project. **Does this server support listing segments and environments?** Yes, the server includes `list_segments` and `list_environments` tools, allowing you to audit your Unleash configuration and targeting rules easily.