Recommendations

TasteRay Skills

Skills that give AI agents the ability to understand people — what drives them, what they value, and what they'll love next.

Built on the open [Agent Skills](https://agentskills.io) standard. Works with [Claude Code](https://claude.ai/code), [Claude Cowork](https://claude.com/blog/cowork-research-preview), [OpenClaw](https://openclaw.ai/), [Cursor](https://cursor.com/), [OpenAI Codex](https://developers.openai.com/codex), [Gemini CLI](https://geminicli.com), [Windsurf](https://codeium.com/windsurf), [GitHub Copilot](https://github.com/features/copilot), [Goose](https://block.github.io/goose/), [Roo Code](https://roocode.com), and [other compatible agents](https://agentskills.io/home).

What These Skills Do

TasteRay skills turn your AI agent into a perceptive conversationalist. Instead of asking blunt survey questions, the agent uses research-backed techniques to understand users through natural dialogue — then applies that understanding to deliver genuinely personalized recommendations.

The two skills work together:

undefined

Elicitation

Deep psychological profiling through patient, research-backed conversation. Your agent learns to uncover values, motivations, and formative experiences — not by interrogating, but by creating space for authentic self-disclosure.

**What it enables:**

undefined

**Grounded in research from:**

undefined

...