菜单
Specialized skill for synthesizing qualitative user research into actionable insights. Analyzes interview transcripts, extracts patterns and themes, identifies pain points, creates affinity diagrams, and generates persona attributes from research data.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
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。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/a5c-ai/babysitter --skill user-research-synthesis命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Reference for querying the Atlas knowledge graph through its MCP tools — the SECONDARY enrichment/comparison layer that adds best-practice context to systems you have ALREADY scanned from your real sources (`az`, repos, dirs). Use when you need to look up nodes, edges, kinds, clusters, stats, or wiki pages in Atlas to compare against your real inventory. (atlas graph, query atlas, atlas mcp, search the graph, graph neighbors, atlas record, atlas kinds, enrichment layer)
Atlas turns your STATED NEED into a real systems atlas by SCANNING your actual sources (Azure via `az`, git repos, local dirs) and process/data mining them, THEN enriching against the Atlas knowledge graph. Use this skill when asked to inventory/map your real systems, scan your cloud + repos + directories, mine the real processes or data they contain, or collect their real constraints/gotchas. (atlas, scan my systems, inventory our azure account, map my repos, real systems atlas, process mining, data mining, collect nuances, system discovery)
This skill should be used when the user asks to "find skills in the wild", "assimilate popular workflows", "discover SKILL.md files in repos", "research external skills", "find workflow patterns", "survey the skill landscape", "what skills exist out there", or wants to investigate public repositories for extractable processes, babysitter plugins, and reusable procedural insights. Searches GitHub for SKILL.md files, classifies repos by archetype, and maintains structured research under docs/reference-repos/.
基于 SOC 职业分类
正在显示 SKILL.md
| name | user-research-synthesis |
| description | Specialized skill for synthesizing qualitative user research into actionable insights. Analyzes interview transcripts, extracts patterns and themes, identifies pain points, creates affinity diagrams, and generates persona attributes from research data. |
| allowed-tools | Read, Grep, Write, Bash, Edit, Glob |
| graph | {"domains":["domain:software-engineering"],"specializations":["specialization:product-management"],"skillAreas":["skill-area:user-research","skill-area:product-discovery"],"roles":["role:product-manager","role:product-analyst"],"workflows":["workflow:product-discovery","workflow:competitive-analysis"]} |
Synthesize qualitative user research data into actionable product insights with thematic analysis and evidence-based recommendations.
This skill provides comprehensive capabilities for transforming raw user research data into structured insights. It supports interview transcript analysis, survey response synthesis, support ticket mining, and cross-source research aggregation.
Supported formats:
- Interview transcripts (.txt, .md, .docx)
- Survey exports (.csv, .xlsx)
- Support ticket exports (.csv, .json)
- User feedback logs (.json, .csv)
## Analysis Framework
### Step 1: Initial Coding
For each transcript:
1. Read through completely for context
2. Highlight significant statements
3. Apply initial codes (open coding)
4. Note participant metadata
### Step 2: Pattern Recognition
Across transcripts:
1. Group similar codes
2. Identify recurring themes
3. Note frequency of mentions
4. Track contradicting evidence
### Step 3: Insight Generation
For each theme:
1. Define the insight clearly
2. List supporting evidence (3+ quotes)
3. Assess confidence level
4. Note actionable implications
{
"transcript_id": "INT-001",
"participant": {
"id": "P001",
"segment": "power_user",
"tenure": "2_years"
},
"findings": [
{
"code": "onboarding_friction",
"theme": "First-time experience",
"quote": "I had no idea where to start...",
"timestamp": "00:12:34",
"sentiment": "negative",
"intensity": "high"
}
],
"summary": {
"key_pain_points": [],
"unmet_needs": [],
"positive_experiences": [],
"feature_requests": []
}
}
## Affinity Diagram Process
### 1. Capture Observations
- One observation per note
- Include source attribution
- Maintain original language
### 2. Group Bottom-Up
- Cluster similar observations
- Name each cluster
- Create hierarchy of clusters
### 3. Output Format
# Theme: [Theme Name]
## Subtheme: [Subtheme Name]
- Observation 1 (P001, INT-001)
- Observation 2 (P003, INT-003)
- Observation 3 (P007, INT-007)
### Evidence Strength
- Strong: 5+ supporting observations
- Moderate: 3-4 supporting observations
- Emerging: 2 supporting observations
## Insight Template
### Insight ID: INS-001
**Statement**: [Clear, actionable insight statement]
**Theme**: [Parent theme]
**Confidence**: [High/Medium/Low]
**Evidence Count**: [Number of supporting data points]
### Supporting Evidence
| Source | Quote | Participant |
|--------|-------|-------------|
| INT-001 | "..." | P001 |
| INT-003 | "..." | P003 |
| SUR-045 | "..." | R045 |
### Implications
- Product: [Product implications]
- Design: [Design implications]
- Engineering: [Technical considerations]
### Recommendations
1. [Specific recommendation]
2. [Specific recommendation]
### Contradicting Evidence
- [Note any contradicting findings]
const researchSynthesisTask = defineTask({
name: 'research-synthesis',
description: 'Synthesize user research into actionable insights',
inputs: {
transcriptPaths: { type: 'array', required: true },
researchQuestion: { type: 'string', required: true },
outputFormat: { type: 'string', default: 'markdown' },
minEvidenceThreshold: { type: 'number', default: 3 }
},
outputs: {
themes: { type: 'array' },
insights: { type: 'array' },
personas: { type: 'array' },
recommendations: { type: 'array' }
},
async run(inputs, taskCtx) {
return {
kind: 'skill',
title: 'Synthesize user research findings',
skill: {
name: 'user-research-synthesis',
context: {
operation: 'full_synthesis',
transcriptPaths: inputs.transcriptPaths,
researchQuestion: inputs.researchQuestion,
outputFormat: inputs.outputFormat,
minEvidenceThreshold: inputs.minEvidenceThreshold
}
},
io: {
inputJsonPath: `tasks/${taskCtx.effectId}/input.json`,
outputJsonPath: `tasks/${taskCtx.effectId}/result.json`
}
};
}
});
## JTBD Statement Format
When [situation/context],
I want to [motivation/goal],
So I can [expected outcome].
### Extraction Process
1. Identify triggering situations in transcripts
2. Extract stated and unstated motivations
3. Map to desired outcomes
4. Categorize: Functional, Emotional, Social jobs
| Severity | Frequency | Impact | Priority |
|---|---|---|---|
| Critical | 80%+ users | Blocks core task | P0 |
| High | 50-80% users | Significant friction | P1 |
| Medium | 25-50% users | Noticeable issue | P2 |
| Low | <25% users | Minor annoyance | P3 |
# Research Synthesis Report
## Executive Summary
[2-3 sentence overview]
## Research Methodology
- **Method**: [Interviews/Surveys/etc.]
- **Participants**: [N participants]
- **Duration**: [Date range]
- **Research Questions**: [Key questions]
## Key Themes
### Theme 1: [Name]
[Description and evidence]
### Theme 2: [Name]
[Description and evidence]
## Top Insights
1. **Insight**: [Statement]
- Evidence: [Count]
- Confidence: [Level]
- Recommendation: [Action]
## Persona Implications
[How findings affect personas]
## Recommended Actions
1. [Action item]
2. [Action item]
## Appendix
- Full coding scheme
- Participant demographics
- Raw data references