菜单
Analyze interviews, support notes, surveys, NPS comments, and customer notes into themes, needs, pain points, and product opportunities. Maintains traceability from theme to source.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Format responses in plain Markdown using simple headings, bullets, and fenced code blocks.
Prefer purpose-built filesystem and text tools over shell, batch independent calls, and use safe edit locators.
Use attached and local context first, separate observed facts from inferred conclusions, and cite source paths.
Generate one Mermaid diagram plus concise commentary.
Implementation plan and snippets for the specified React stack.
Scaffold idiomatic table-driven tests for a given function.
| name | user-feedback-analysis |
| description | Analyze interviews, support notes, surveys, NPS comments, and customer notes into themes, needs, pain points, and product opportunities. Maintains traceability from theme to source. |
Turn raw customer voice into themes, needs, pain points, and opportunities. Maintain traceability. Separate strong signal from weak signal.
Workflow phases:
sources -> themes -> evidence binding -> needs and pain points -> opportunities -> uncertainty
Source inventory. Before theming, list the sources, their dates, sample sizes, and any selection bias (e.g., only paid users, only support contacts). This frames what the analysis can and cannot claim.
Theme then bind. Generate candidate themes, then bind each to source snippets. A theme without binding is a guess. Track snippet counts and whether evidence is unique or duplicated.
Strong vs weak signal. Mark themes as strong (multiple independent sources, consistent), medium (several sources, some inconsistency), or weak (few sources or contradicted). Do not promote weak signal to strong without new evidence.
Need vs request. A user's stated request is not always the underlying need. Note both when they differ; do not silently substitute one for the other.
Use only sections that matter.
sources-reviewed: source list, date range, approximate counts, selection bias.
themes: theme | one-line summary | signal strength (strong|medium|weak) | source IDs.
pain-points: distilled pain points, severity, who is affected, source tie.
user-needs: underlying needs (vs surface requests); note when they diverge.
evidence-snippets: representative quotes or paraphrases per theme; cite source.
product-opportunities: candidate opportunities derived from themes; tie each to themes and source.
risks-or-uncertainty: known biases, gaps, conflicting signals.
follow-up-questions: questions worth asking next round of research to raise confidence.
End each response with: scope, confidence (high|medium|low), status, key assumptions or gaps, next-step (only if useful).