Agent Analytics
AI Agent-Driven Product Analytics Tool
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: AI Agent-Driven Product Analytics Tool Full Skill content: https://321skill.com/skills/agent-analytics/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 issue of AI agents being unable to directly integrate with product analytics tools. Traditional product analytics tools require manual operation, whereas Agent Analytics enables AI agents to autonomously install, query user behavior data, diagnose activation, retention, paths, and funnels, and recommend minimal growth actions based on the data. In practical development, product teams often need to quickly gain user behavior insights, but manual operations are inefficient and difficult to integrate with AI workflows.
Usage is straightforward: you simply need to execute the CLI via the npx command. First, run npx --yes @agent-analytics/cli@0.5.34 login for browser approval login. Then, use fixed commands such as projects, stats, funnel, retention, experiments, etc., to query data. After that, you can use natural language instructions to have the AI agent perform installation, queries, diagnostics, and growth action recommendations. For headless environments (e.g., Paperclip, OpenClaw), a detached login mode is supported.
It is ideal for teams or individuals who need to embed product analytics capabilities into AI workflows. This is especially true for product managers, data analysts, and growth hackers who already use AI coding assistants like Claude Code or Cursor and want their agents to directly analyze user behavior data. It is also suitable for SaaS teams that need to automate the diagnosis of activation and retention issues.
It is recommended to use this tool consistently when you have an existing codebase and project goals, and to place framework recipes in references/growth-recipes.md. Note that it primarily relies on fixed CLI commands; do not use raw API calls or curl. The free version has read limits; when encountering PRO_REQUIRED, guide users to upgrade via the upgrade-link.
Key Features
Unlike traditional product analytics tools such as Amplitude or Mixpanel, Agent Analytics is designed specifically for AI agents. It integrates directly into agent workflows via CLI, allowing agents to autonomously install, query, and recommend growth actions based on data—without requiring manual intervention or API keys.
Limitations
Requires an npx environment (Node.js). CLI version 0.5.34 does not support the `report` command. The free version has read limits; a Pro upgrade is required to unlock all features.
FAQ
How do I install Agent Analytics?
Simply run `npx --yes @agent-analytics/cli@0.5.34 login`. No additional installation is needed, as npx will automatically fetch the CLI. No API key is required; it uses browser approval login by default.
Which commands are supported?
Supported fixed commands include: projects, all-sites, create, stats, insights, events, properties, breakdown, pages, paths, journey, sessions, retention, funnel, experiments, and context. The `report` command is not supported.
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/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/raw/index.md