summarize-interview
by phurynsummarize-interview turns a customer interview transcript into a structured discovery summary with JTBD, current solution, satisfaction signals, key insights, and action items. Use it for interview recordings, transcript cleanup, and concise summaries for product teams, researchers, founders, or Data Analysis workflows.
This skill scores 78/100, which means it is a solid but not premium listing: directory users can reasonably install it if they need structured interview summaries, but they should expect a lightweight skill with limited ecosystem support. The repository gives enough workflow guidance to trigger and use it with less guesswork than a generic prompt.
- Clear use case and trigger text for customer interview transcripts, discovery interviews, and interview summaries.
- Provides a concrete output template covering JTBD, satisfaction signals, key insights, and action items.
- Includes practical instructions to read the full transcript first and use simple language, which improves agent execution.
- No supporting scripts, references, or resources, so there is little external validation or advanced automation guidance.
- The skill appears narrowly scoped to transcript summarization and may need adaptation for non-standard interview formats or incomplete transcripts.
Overview of summarize-interview skill
The summarize-interview skill turns a raw customer interview transcript into a structured discovery summary with JTBD, satisfaction signals, key insights, and action items. It is best for product teams, researchers, founders, and analysts who need a fast, consistent way to convert interview notes or transcripts into something decision-ready. If you are doing product discovery, synthesis after a call, or a first-pass recap for stakeholders, summarize-interview reduces the gap between “we recorded something” and “we know what it means.”
What this summarize-interview skill is for
The skill is designed to summarize one interview at a time, not to run thematic analysis across a whole research program. Its value is in structure: it helps preserve the customer’s current solution, pain points, perceived value, and follow-up actions in a format that is easier to scan than a freeform summary.
Best fit for discovery and note-taking
Use summarize-interview when you already have a transcript and want a clean output for product discovery, user research, or interview documentation. It is especially useful when teams need a repeatable summary format across many interviews instead of relying on each note-taker’s style.
What makes it different
The skill emphasizes job-to-be-done language and simple wording, which is useful when summaries will be shared beyond the research team. It also asks for missing information to be marked clearly, which helps avoid fabricated detail and makes gaps visible instead of silently filling them in.
How to Use summarize-interview skill
Install summarize-interview
Install with:
npx skills add phuryn/pm-skills --skill summarize-interview
That is the summarize-interview install path used in the repository. After install, treat the skill as a transcript-to-summary workflow, not a general writing assistant.
Give it the right input
The summarize-interview usage pattern expects a full interview transcript or an attached file that contains the transcript. It works best when you provide speaker names, timestamps if available, and any context that clarifies the interview goal. A weak prompt like “summarize this” usually gives a weaker summary than “summarize this interview for product discovery; focus on JTBD, current workaround, pain points, and action items.”
Recommended workflow
Start by reading SKILL.md, then inspect any nearby supporting instructions in the repo tree. In this repository, the summarize-interview guide is intentionally compact, so the main value comes from understanding the template and then adapting it to your own data source.
A practical workflow is:
- Attach or paste the full transcript.
- State the audience if needed, such as product, design, or customer success.
- Ask for the summary in the skill’s template style.
- Review missing fields and unclear statements before sharing the output.
Where it helps most
For Data Analysis work, summarize-interview is useful when interview findings need to be moved into a structured artifact for later comparison, tagging, or stakeholder review. It is not a statistics tool and does not replace coding or thematic synthesis, but it gives analysts a more consistent starting point than raw notes.
summarize-interview skill FAQ
Is summarize-interview just a generic prompt?
No. A generic prompt can summarize text, but the summarize-interview skill is narrower and more decision-oriented. It pushes the output toward discovery-relevant fields like current solution, problems, satisfaction, and action items, which makes the result more useful for product teams.
When should I not use summarize-interview?
Do not use it when you need cross-interview synthesis, survey analysis, or a polished meeting recap that ignores discovery structure. It is also a poor fit if the source material is incomplete and you need a high-confidence factual report; the skill is better when the transcript is complete enough to support a structured summary.
Is summarize-interview beginner friendly?
Yes. The summarize-interview skill is beginner friendly because the output template is simple and the instructions are direct. The main requirement is that you provide a complete transcript and avoid asking it to infer details that are not actually present.
Does it fit Data Analysis workflows?
Yes, especially when the goal is to standardize interview data before analysis. For summarize-interview for Data Analysis, the key is consistency: use the same prompt framing and same transcript quality so summaries can be compared later without rework.
How to Improve summarize-interview skill
Provide fuller transcripts and clearer context
The biggest quality gain comes from transcript completeness. Include speaker labels, customer role, product context, and the interview objective. For example, “Summarize this 45-minute interview with a payroll admin about switching vendors; focus on pains, current workaround, and buying signals” gives better output than a bare transcript dump.
Ask for the summary shape you need
If your team cares about roadmap decisions, say so. If you need a research artifact, say that the summary should preserve quotes, dissatisfaction cues, and action items. The summarize-interview skill performs best when the prompt tells it what to emphasize rather than assuming one template fits every downstream use.
Watch for common failure modes
The most common issue is over-compression: important nuance can disappear if the transcript is vague or noisy. Another issue is over-inference, where an assistant may make the customer sound more certain than they were. Reduce both by asking the model to use “-” for missing fields and to keep wording simple and faithful to the transcript.
Iterate on the first output
Use the first pass to spot gaps, then refine with a second prompt that targets those gaps only. For example: “Re-run the summarize-interview output and strengthen the action items, add exact customer phrasing for the top pain point, and clarify the current solution.” That kind of iteration usually improves the summary more than asking for a longer rewrite.