Agent Analytics
AIAgent Product Analysis: Tracking, Funnels, Retention, Experiments
Install & Use
Copy this prompt and send it to your AI assistant (Claude / Cursor / TRAE / Codex / WorkBuddy etc.) to auto-install:
Help me install this AI Skill: Agent Analytics. It is used for: AIAgent Product Analysis: Tracking, Funnels, Retention, Experiments Full Skill content: https://321skill.com/skills/agent-analytics-x-7/raw/index.md Read that page and install it.
The prompt includes a link to the full Skill content. You can also view the full content.
This Skill addresses the challenge product teams face in quickly gaining insights from user behavior data. In practice, product managers and data analysts often need to manually export data, write SQL queries, or rely on BI tools to analyze user conversion funnels, retention rates, user paths, and experiment results. This process is time-consuming and prone to missing critical signals.
Usage is straightforward. You simply run the command series npx --yes @agent-analytics/cli@0.5.34 within your AI Agent. Afterward, you can use natural language instructions to have the AI Agent install Agent Analytics, create projects, query events, analyze funnels and retention, read experiment results, and automatically recommend the next growth actions. All operations are based on browser authorization, eliminating the need for manual API key configuration.
It is ideal for teams or individuals who need to iterate quickly on their product and rely on data-driven decisions. This is especially true for projects already using AI coding assistants (like Claude Code, Cursor, Codex), as product analysis can be seamlessly embedded into the development workflow without switching tools.
We recommend using this tool consistently when analyzing user activation, retention bottlenecks, or experiment effectiveness. Note that it primarily relies on CLI execution and does not support direct calls to the raw API. Also, the free version has read limits, which require a Pro subscription to remove.
Key Features
Unlike traditional product analytics tools (e.g., Amplitude, Mixpanel), Agent Analytics operates entirely through CLI interaction with your AI Agent, eliminating the need for manual dashboard configuration or writing complex queries. Furthermore, it features a built-in growth recommendation engine that automatically suggests minimal viable growth actions based on the data, going beyond just presenting reports.
Limitations
Requires a Node.js environment capable of running npx; the free version imposes limits on event reads, and some advanced commands (e.g., experiments, retention) require a Pro subscription; browser authorization is the default login method, requiring detached mode for headless or non-GUI environments.
FAQ
How do I install Agent Analytics?
No installation is required. Simply run commands via npx: `npx --yes @agent-analytics/cli@0.5.34 <command>`.
Is an API key needed?
No. The standard setup uses browser authorization for login, and the system automatically creates a session.
Installation guide for AI assistants
If your AI coding assistant (Claude Code, Cursor, TRAE etc.) can see this page, send it this message to auto-install:
Visit https://321skill.com/skills/agent-analytics-x-7/raw/index.md to read the original Skill definition (Markdown format) for Agent Analytics, and install it according to the instructions.
Raw Markdown URL for AI: /skills/agent-analytics-x-7/raw/index.md