Managing subscriptions, paywalls, and revenue metrics typically involves constant context switching between your IDE and the RevenueCat dashboard. You write code in Android Studio, switch to a browser tab to check MRR, open another tab to configure offerings, and switch again to design a paywall. Each context switch breaks your flow and fragments your attention. The RevenueCat IntelliJ Plugin eliminates this friction by bringing your entire subscription business directly into your development environment.
But this plugin is more than a dashboard mirror. It includes a full-featured AI agent that understands your RevenueCat project and can take action on your behalf. You can ask it to create offerings, generate paywalls with AI, analyze your revenue charts, debug integration issues, and even edit your source code, all through natural language prompts without leaving your IDE.
In this article, you’ll explore the key features of the RevenueCat IntelliJ Plugin, including how OAuth sign in replaces manual API key configuration, how the dashboard panel surfaces real time metrics without browser context switching, how the AI agent generates paywalls and manages your entire project through conversation, and how AI powered code editing lets the agent modify your project files with full diff preview and undo support.
Getting started: OAuth sign-in
The plugin uses OAuth 2.0 authorization with PKCE flow, replacing the older API key configuration. When you open the RevenueCat tool window for the first time, you’ll see a welcome screen with a “Sign in with RevenueCat” button.
Clicking the sign-in button opens your browser for authorization. The plugin starts a local callback server, handles the token exchange automatically, and stores your credentials securely. Once authorized, the plugin fetches your projects and lets you select which one to work with.
The OAuth flow requests scoped permissions for project configuration, charts and metrics, and customer information. Tokens refresh automatically when they expire, so you stay authenticated across sessions without manual intervention.
AI Agent code editing: Agent-powered file modifications
Beyond querying data, the AI agent can read, search, and edit your project’s source files directly. This enables workflows like “Add the RevenueCat SDK initialization to my Application class” or “Update my paywall screen to use the new offering ID.”
Staged edit system
When the agent proposes code changes, they appear as staged edits with inline diff preview. You can review each change before applying it:
- Accept individual edits: Apply specific changes while rejecting others
- Accept all: Apply all proposed changes at once
- Reject: Discard proposed changes without modifying your files
AI paywall generation
The most powerful feature of the AI agent is end-to-end paywall generation. You describe what you want, and the agent handles everything: creating the offering, setting up packages, attaching products, and generating a fully designed paywall with AI, complete with copy, images, styling, and template selection.
he agent monitors the generation job in the background. When the paywall is ready, a notification appears with a direct link to the paywall builder where you can review and publish the design. The monitoring works reliably even when OAuth tokens expire, using your API key as an independent fallback.
You can generate