cookiy-ai/user-research-skill

What it solves

It bridges the gap between AI agents and real-world human feedback. Instead of relying solely on training data or synthetic assumptions, it allows AI agents to plan, execute, and analyze actual user research—including qualitative interviews and quantitative surveys—directly within the agent's interface.

How it works

This project acts as a "skill" or plugin that can be integrated into AI agents (such as Claude Code, Cursor, or Codex). It provides the agent with a set of capabilities to generate research plans and interview guides, synthesize raw transcripts into structured reports with personas and codebooks, and connect to the Cookiy AI platform to recruit and moderate interviews or surveys with real or synthetic participants.

Who it’s for

It is designed for product managers, UX researchers, and developers who use AI agents to build products and want to integrate evidence-backed human insights into their development workflow.

Highlights

• End-to-End Research: Handles everything from planning and screening to data collection and final synthesis.
• Hybrid Methodology: Supports both qualitative (AI-moderated interviews) and quantitative (multi-language surveys with conditional logic).
• Agent Integration: Works as a plugin for various AI platforms, allowing the agent to trigger research tasks via semantic matching or explicit commands.
• Synthesis Pipeline: Converts raw interview transcripts into evidence-backed reports featuring prioritized findings.