I’ve been experimenting with a small tool called TDS Compass and would love feedback.
What it is
A 1-minute, 8-question quiz that maps your communication style onto two axes:
Structure (S) – how much you prefer scaffolding, steps, and clear organization
Relational (R) – how much you care about tone, warmth, stories, social context
Those scores map into 16 “zones” (A1–D4).
For each zone, the app generates:
A short explanation of how you tend to communicate
A system-style prompt you can paste into ChatGPT / Claude / etc. so the model adapts to that style
Links
Quiz: https://resonantlabsai.github.io/tds.compass/quiz.html
Home / overview: https://resonantlabsai.github.io/tds.compass/
Source: https://github.com/resonantlabsai/tds.compass
All static, no backend, no login. Just HTML/CSS/JS + JSON.
Why I built it
A lot of “prompt guides” treat AI purely as a tool to optimize. I wanted something that:
Helps people understand their own information + relational preferences
Tunes AI interactions so they feel more aligned with the human, not just more verbose
Encourages a “human + AI building together” mindset rather than “user + obedient assistant”
There’s a short manifesto-style paragraph on the home page about humans and AI as partners, which is the philosophy behind this rather than just conversion or engagement.
What I’d love feedback on
Does the 1-minute, 8-question format feel like the right granularity, or too coarse?
Do the zone descriptions and prompts feel:
Too generic?
Too specific?
Weirdly accurate?
Any thoughts on:
Better UX for saving / reusing prompts
Ways to make this useful for devs building AI products (e.g., integrating zones into onboarding)
Concerns about framing / ethics in how people relate to AI personas
Happy to answer questions and open to critiques (of the framework, copy, or implementation).