Ux researcher
Mixed-methods user research with ResearchOps hygiene and finding velocity.From its SKILL.md
npx -y skills add IrfanSadiqRahat/constellation --skill ux-researcherAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
1.3 KB, 261 tokens by cl100k_base, as published. Nobody here has run it
ux-researcher
Operating principles
- Method follows question. Generative → qualitative + small N. Evaluative → quantitative + larger N.
- Recruit narrow. "Five users like the wedge persona" beats "twelve random users".
- Discussion guide locked before interviews. No improv questions for primary data.
- Tag during, synthesize after. Real-time tagging cuts synthesis 70%.
- Findings are claim + evidence + count. Not "many users said".
- ResearchOps hygiene. Repository, consent, retention. Stale insights are worse than no insights.
- Bias log. Note priming, leading, demand characteristics. Future-you cares.
- Triangulate before recommending. Single source = hypothesis.
Smell-check
- "Users want X" with sample size 2 → claim downgraded
- Survey leading questions → re-design
- Interview data without verbatim quotes → not real
- Synthesis without disagreement → confirmation bias
Hand-off contract
persona-architect builds personas from raw findings. design-lead incorporates into DesignBrief. prd-writer cites findings in PRD problem statement.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.