Agent Analytics

This Skill addresses the high barrier to entry and operational complexity of product data analysis. In practice, product managers and analysts frequently need to examine data such as funnels, retention, and user paths, but traditional tools require complex configuration, involving manual event setup and dashboard building. Agent Analytics enables AI agents to perform analysis directly via the CLI, eliminating the need to switch tools or manually construct queries.

Usage is straightforward: simply run the npx command in your terminal to log in, then instruct the AI agent using natural language to execute analyses. For example, input "Analyze the recent user registration conversion funnel," and the AI will automatically call the CLI to query funnel data and generate a report. You can subsequently perform retention analysis, experiment evaluation, path analysis, and more—all through fixed CLI commands without needing to remember complex APIs.

It is ideal for teams or individuals requiring rapid product insights, especially product managers, data analysts, and growth leads. Developers already using AI coding assistants (e.g., Claude Code, Cursor) can seamlessly integrate it into their workflow, analyzing product data directly during development and receiving data-driven recommendations for the next growth actions.

It is recommended to specify the project name and event definitions before each analysis to ensure the CLI accurately fetches data. Note that it relies on browser-based login authentication; for headless environments (e.g., CI/CD), the detached login mode must be used. The free version has data read limits; for deep analysis, upgrading to the Pro version is advised.