ux-research
Skill1 starsChanged 4 months agoMITView on GitHub
UX research and user interview analysis. Use when the user shares interviews, transcripts, survey responses, support tickets, or user feedback to analyze — extracts insights, pain points, and design implications, and documents them in the issue's spec folder.
16 other projects use a version of this.
Adopted by
Most recently taken up by
- lengrufeng2008/dsh-skill-librarytheir own version, first seen Nov 15, 2026
- adymire/CraftLayertheir own version, first seen Nov 15, 2026
- califa/skillstheir own version, first seen Nov 15, 2026
- redhoram/claude-landing-templatetheir own version, first seen Nov 7, 2026
- its-ash/skillstheir own version, first seen Nov 1, 2026
- omspradippatil/Claude-Skillstheir own version, first seen Nov 1, 2026
Other versions16 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 16
- nodetool-ai/nodetool★ 637same name
- cuellarfr/design-skills★ 94same name
- felvieira/claude-skills-fv★ 28same name
- Gustavosilveira23/claude-design-skills★ 9same name
- justinqwerty/design-skills★ 9same name
- pedroromeroluna/ai-first-product-skills★ 5same name
- omspradippatil/Claude-Skills★ 3same name
- califa/skills★ 2same name
- KMaruthi2002/universal-llm-skills★ 1same name
- redhoram/claude-landing-template★ 1same name
6 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository9 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- backend-architectureSkill
- competitive-researchSkill
- figma-to-codeSkill
- frontendSkill
- lead-generationSkill
- seoSkill
- spec-writingSkill
- web-designSkill
The file
We split this skill into the rules it gives an agent
5 rules. 0 are common, 5 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 14 edited versions we read keep these.
5
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
All research documents live inside the spec folder of the related issue:
- Distinctive: line 24
If the research is done before an issue exists, create the issue first so the folder has its number.
- Distinctive: line 29
When the user shares one or more interviews or transcripts:
- Distinctive: line 62
Every finding must land somewhere concrete:
- Distinctive: line 72
Never invent data: if the transcripts don't support a claim, say so.
“Common” is measured: kept by at least 6 in 10 of the 14 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
UX Research
Turn raw user input (interviews, transcripts, surveys, feedback) into actionable design and product decisions.
Output Location (strict)
All research documents live inside the spec folder of the related issue:
/specs/<issue-number>-<slug>/
spec.md
research/
interviews.md # interview analysis
personas.md # personas (when enough data exists)
insights.md # synthesized findings + design implications- Written in English, like all documentation, even if the interviews are in another language (quote original language verbatim, translate the insight).
- If the research is done before an issue exists, create the issue first so the folder has its number.
- Reference the research files from
spec.md(section 2, Problem / Goal).
Interview Analysis Workflow
When the user shares one or more interviews or transcripts:
- Read everything first — no conclusions until all material is read.
- Code the data: tag each relevant quote with a theme (pain point, goal, workaround, objection, delight, vocabulary).
- Separate observation from interpretation: what the person SAID vs what it MEANS. Keep verbatim quotes as evidence — never paraphrase into the analysis without the original quote.
- Weight behavior over opinion: "I would pay for that" is weak; "I currently pay for X / I built a spreadsheet to solve this" is strong.
- Synthesize across interviews: a theme needs 2+ independent sources to be a finding; single mentions go to an "isolated signals" list.
interviews.md structure
# Interview Analysis: <topic>
> Issue: #<number> · Sources: N interviews · Date: YYYY-MM-DD
## Method & Sources
Who was interviewed, how many, context, limitations of the data.
## Themes
### T1: <Theme name> (N/M interviewees)
- Finding: <what we learned>
- Evidence: "<verbatim quote>" — P3
- Design implication: <what this means for the site/feature>
## Pain Points (ranked by frequency × severity)
## Goals & Jobs-to-be-Done
## Vocabulary (words users actually use — feed into copy and SEO keywords)
## Objections & Anxieties (feed into trust signals and FAQ)
## Isolated Signals (1 mention, watch for recurrence)
## Open Questions (what we still don't know)From Insights to Design Decisions
Every finding must land somewhere concrete:
- Pain points → features, content, or fixes (reference them in spec.md's solution).
- User vocabulary → page copy, headlines, CTA labels, and SEO keyword targets (users search with their words, not ours).
- Objections/anxieties → trust signals, guarantees, FAQ entries near conversion points.
- Goals/JTBD → page structure and information architecture priorities.
- Traceability: in
spec.md, cite the finding ID (e.g. "T3") that justifies each significant decision.
Analysis Discipline
- Never invent data: if the transcripts don't support a claim, say so.
- Note sample bias explicitly (who is missing from the data).
- Distinguish severity: blocker > friction > annoyance > preference.
- Contradictions between interviewees are findings too — document both sides.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

