User interview synthesis
Skill shinzoxD/knackbox/skills/research/user-interview-synthesis
Synthesize user interviews into themes, insights, and opportunities. Use whenever the user pastes interview notes or transcripts and wants themes, affinity mapping, insights, JTBD, or product opportunities from qualitative research — even from messy notes.From its SKILL.md
npx -y skills add shinzoxD/knackbox --skill user-interview-synthesisAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 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.
What its file declares
Copied from the file, not written here
The file declares its own license as Apache-2.0. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
1.9 KB, 320 tokens by cl100k_base, as published. Nobody here has run it
User Interview Synthesis
Qual synthesis turns anecdotes into decision-useful patterns without erasing dissent. Tag evidence. Separate observation from interpretation.
Workflow
- Note research goal and participant mix (if known).
- Extract atomic observations and quotes (paraphrase ok; mark exact quotes).
- Cluster into themes; count how many participants support each (n).
- Write insights as "force of nature" statements with evidence.
- Opportunities and recommendations with confidence.
- Gaps and follow-ups (who else to interview, what to validate quantitatively).
Output format
## Synthesis: <study name>
**Participants:** n=…
**Goal:** …
### Themes
#### Theme title (n=k/N)
- Evidence bullets / quotes
- Counterexamples
### Insights
1. …
### Opportunities
| Opportunity | Evidence | Confidence | Effort guess |
### Divergent perspectives
…
### Method notes & limitations
…
Rules
- Never invent quotes or participants.
- One loud user ≠ a theme; show counts or caution.
- Keep raw pain language; do not sanitize into corporate mush too early.
- Separate what users do vs what they say they want.
- Label speculation clearly.
- Privacy: strip PII not needed for product decisions.
Edge cases
- n=1–2: micro-synthesis + strong caution.
- Mixed segments: theme within segment first.
- Stakeholder wants only highlights: still attach evidence links/anchors.
What ships with it: 1 file
1.4 KB alongside SKILL.md
benchmarks/
- prompts.json1.4 KB