This project offers an AI agent skill for end-to-end user research, connecting AI platforms directly with human insights. It enables AI agents to plan, conduct, and synthesize qualitative (interviews) and quantitative (surveys) studies without leaving the conversation. The skill benefits developers and researchers by streamlining the user research lifecycle and providing AI agents access to real human opinions and decision-making processes.

How It Works

The skill functions as a "human layer" for AI agents, providing four core capabilities: "Plan" (generating research designs, screeners, guides), "Synthesize" (transforming transcripts into reports with personas and findings), "Qual" (AI-moderated interviews with real/synthetic participants via Cookiy AI), and "Quant" (multi-language surveys with recruitment and results analysis via Cookiy AI). It integrates with various AI agents, responding to explicit commands or semantic understanding of research goals.

Quick Start & Requirements

Installation varies by agent:

• Claude Cowork/Code Desktop: Via Plugin Marketplace (cookiy-ai/user-research-skill). Requires network egress and allowing s-api.cookiy.ai or all domains.
• Claude Chat Desktop: Download ZIP, upload skill. Network configuration is identical.
• Claude Code Terminal: Use bash /plugin marketplace add cookiy-ai/user-research-skill, /plugin install user-research@cookiy-ai, /reload-plugins. Auto-update is optional.
• Codex/Cursor/OpenClaw/Other Agents: Execute npx cookiy-ai or follow agent-specific manual installation guides. A key requirement is enabling network access for agent communication with Cookiy AI services.

Highlighted Details

• Full user research lifecycle support: planning, qualitative/quantitative studies, and synthesis.
• Integration with Claude, Codex, Cursor, and OpenClaw AI agents.
• AI-moderated interviews using real or synthetic participants.
• Multi-language surveys with conditional logic and respondent recruitment.
• Automated synthesis pipeline for actionable insights and personas.

Maintenance & Community

No specific details on contributors, sponsorships, community channels, or roadmaps are provided in the README.

Licensing & Compatibility

Released under the MIT License. While permissive for commercial use, core functionality depends on "Cookiy AI" services, which may have separate terms or costs not detailed here.

Limitations & Caveats

The README does not explicitly list limitations. However, full execution of qualitative and quantitative study features relies on integration with the external "Cookiy AI" platform services.