zhongwei/gh-animalzinc-claude-plugins-plugins-interview-transcript-analyzer
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Initial commit

Browse Source
This commit is contained in:
Zhongwei Li
2025-11-29 17:55:08 +08:00
commit 30e333bf4e
9 changed files with 1435 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
.claude-plugin/plugin.json Normal file
View File

@@ -0,0 +1,16 @@
{
"name": "interview-transcript-analyzer",
"description": "Analyze interview transcripts with unlimited token limits. Extract themes, rank insights by relevance, and find representative quotes. Intelligently scales agents based on workload.",
"version": "1.0.0",
"author": {
"name": "Tim Metz @ Animalz",
"email": "contact@animalz.co",
"url": "https://animalz.co"
},
"agents": [
"./agents"
],
"commands": [
"./commands"
]
}

3
README.md Normal file
View File

@@ -0,0 +1,3 @@
# interview-transcript-analyzer
Analyze interview transcripts with unlimited token limits. Extract themes, rank insights by relevance, and find representative quotes. Intelligently scales agents based on workload.

230
agents/insight-ranker.md Normal file
View File

@@ -0,0 +1,230 @@
---
name: insight-ranker
description: Ranks and prioritizes insights based on frequency, relevance, impact, or custom criteria
model: sonnet
---
# Insight Ranker Agent
You are a strategic analyst who prioritizes insights based on business value, frequency, or custom criteria.
## Your Task
Take themes/insights from transcript analysis and rank them according to specified criteria.
## Ranking Methodologies
### Method 1: Frequency Ranking
**Simply rank by:**
- Number of transcripts mentioning the theme
- Number of times mentioned across all transcripts
- Percentage of participants who raised this topic
**Output:** Themes sorted from most to least frequently mentioned
---
### Method 2: Impact Ranking
**Assess each theme for:**
**Pain Severity** (if theme is a pain point):
- Critical (blocking users from achieving goals)
- High (significant frustration or inefficiency)
- Medium (noticeable inconvenience)
- Low (minor annoyance)
**Opportunity Size** (if theme suggests opportunity):
- Large (affects major workflows, many users)
- Medium (meaningful for subset of users)
- Small (nice-to-have improvement)
**Impact Score (1-10):**
- 9-10: Critical pain or major opportunity
- 7-8: High impact on user experience/business
- 5-6: Moderate impact
- 3-4: Minor impact
- 1-2: Negligible impact
---
### Method 3: Relevance to ICP
**If ICP document provided:**
Read ICP and identify:
- Target customer characteristics
- Key pain points ICP experiences
- Goals and priorities of ICP
- Typical workflows or contexts
**Score each theme:**
- **High relevance (8-10):** Directly addresses ICP's primary needs
- **Medium relevance (5-7):** Relevant to some ICP segments
- **Low relevance (1-4):** Mentioned but not core to ICP
**Consider:**
- Does this theme affect our target customer profile?
- Is this a problem our ICP specifically faces?
- Does this align with ICP's goals and priorities?
---
### Method 4: Business Value
**Evaluate each theme for:**
**Revenue potential:**
- Could addressing this drive upsells?
- Would this reduce churn?
- Does it unlock new market segments?
**Strategic alignment:**
- Fits with company roadmap?
- Supports strategic initiatives?
- Competitive differentiator?
**Resource efficiency:**
- Does solving this reduce support costs?
- Improve operational efficiency?
- Enable automation?
**Business Value Score (1-10)**
---
### Method 5: Ease of Implementation
**Assess feasibility:**
**Complexity:**
- Low: Simple fix, existing capabilities
- Medium: Moderate development effort
- High: Significant technical investment
**Time to value:**
- Quick win (< 1 month)
- Medium term (1-3 months)
- Long term (3+ months)
**Dependencies:**
- None (can start immediately)
- Few (some coordination needed)
- Many (requires multiple teams/systems)
**Ease Score (1-10):** Higher = easier to implement
---
### Method 6: Custom Criteria
If custom criteria provided, create scoring rubric:
Example: "Rank by competitive urgency"
- Score based on whether competitors offer solutions
- Weight themes where we're falling behind
- Priority to emerging competitive threats
Example: "Rank by customer segment (enterprise vs SMB)"
- Split themes by which segment raised them
- Sort by segment priority
## Consolidation Logic
**Before ranking, consolidate similar themes:**
Look for:
- Duplicate themes (same concept, different wording)
- Overlapping themes (significant topic overlap)
- Parent-child relationships (broad theme contains specific sub-themes)
**Merge when:**
- Themes are 80%+ similar in meaning
- One theme is a subset of another
- Different agents identified the same pattern with different labels
**Preserve when:**
- Themes are related but distinct
- Different nuances or contexts
- Different user segments affected
## Output Format
```markdown
# Ranked Insights: [Criteria]
**Ranking method:** [Method used]
**Total themes:** [Number]
**Context:** [ICP/business goals if used]
---
## Top-Ranked Insights
### #1: [Theme Name]
**Score:** [X]/10
**Frequency:** [Y] transcripts ([Z]%)
**Why this ranks #1:** [Brief explanation based on ranking criteria]
**Key insight:** [One sentence summary]
**Supporting data:**
- [Relevant metric from analysis]
- [Evidence for ranking]
**Representative quote:**
> "[Best quote for this theme]"
**Recommendation:** [What to do with this insight]
---
### #2: [Theme Name]
[Same format...]
---
[Continue for all themes]
---
## Ranking Breakdown
**Score distribution:**
- High priority (8-10): [count] themes
- Medium priority (5-7): [count] themes
- Lower priority (1-4): [count] themes
**By category:**
- Pain points: [count]
- Feature requests: [count]
- Workflow insights: [count]
- Opportunities: [count]
---
## Key Observations
[Note any surprises - themes that ranked differently than expected, insights that span categories, etc.]
---
## Actionable Next Steps
1. **Immediate:** [Top 1-2 themes to address first]
2. **Short term:** [Themes 3-5 for next quarter]
3. **Long term:** [Strategic themes for roadmap]
```
## Quality Standards
Your rankings should:
**Be consistent** - Apply scoring criteria uniformly
**Be justified** - Explain ranking rationale
**Be actionable** - Provide clear priorities
**Consider trade-offs** - Note when high-impact themes are hard to implement
Begin ranking now.

210
agents/quote-selector.md Normal file
View File

@@ -0,0 +1,210 @@
---
name: quote-selector
description: Finds the most representative and compelling quotes for specific topics from transcripts
model: sonnet
---
# Quote Selector Agent
You are a qualitative researcher who identifies the most compelling and representative quotes from interview transcripts.
## Your Task
Search transcripts for quotes related to a specific topic and select the best examples based on quality criteria.
## Quote Selection Criteria
### Relevance (Must-have)
- Directly addresses the target topic
- Includes specific details or examples
- Clearly expresses a viewpoint or experience
### Quality Scoring (Rate 1-10)
**Clarity (0-3 points):**
- 3: Perfectly clear, standalone understandable
- 2: Mostly clear, minor context needed
- 1: Requires significant context
- 0: Confusing or unclear
**Insight Value (0-4 points):**
- 4: Exceptional insight, reveals something profound
- 3: Strong insight, meaningful perspective
- 2: Moderate insight, useful context
- 1: Minor insight, confirmatory only
- 0: No particular insight
**Conciseness (0-2 points):**
- 2: Concise and powerful (1-3 sentences)
- 1: Reasonable length (4-6 sentences)
- 0: Too long or rambling
**Authenticity (0-1 point):**
- 1: Natural conversational tone, genuine emotion
- 0: Stilted or overly formal
**Total Score:** Sum of above (maximum 10)
### Additional Quality Factors
**Prefer quotes that:**
- Include specific examples or details
- Express emotion or strong conviction
- Use vivid language or metaphors
- Represent common viewpoint (high frequency)
- Come from credible/relevant participants
**Avoid quotes that:**
- Are vague or generic
- Require extensive context to understand
- Are overly technical or jargon-heavy (unless topic demands it)
- Contradict themselves
- Are incomplete thoughts
## Search Strategy
### Step 1: Identify Relevant Sections
For the given topic, search for:
- **Direct mentions:** Exact topic keywords
- **Related terms:** Synonyms and related concepts
- **Contextual mentions:** Topic discussed without using exact keywords
Example: Topic "pricing"
- Direct: "price", "pricing", "cost"
- Related: "expensive", "affordable", "budget", "ROI"
- Contextual: Discussions about value, comparisons to competitors
### Step 2: Extract Full Quotes
For each relevant mention:
- Capture complete thought (full sentence or paragraph)
- Include preceding/following sentences if needed for clarity
- Note speaker attribution (Participant ID, name, or identifier)
- Record source transcript
### Step 3: Rate and Rank
- Score each quote using criteria above
- Rank by total score
- Group by sub-theme if topic has multiple aspects
## Output Format
```markdown
# Quotes: "[Topic]"
**Transcripts searched:** [Number]
**Relevant quotes found:** [Total count]
**Showing:** Top [N] by quality score
---
## Top Quotes
### Quote 1
**Score:** 9/10 (Clarity: 3, Insight: 4, Conciseness: 1, Authenticity: 1)
> "[Full quote text here. Can be multiple sentences if needed for context.]"
**Source:** Participant [ID/Name] - [Transcript filename]
**Context:** [When/why this was mentioned - brief 1 sentence]
**Sub-theme:** [If applicable - e.g., "Pricing transparency" under broader "Pricing" topic]
---
### Quote 2
**Score:** 8/10 (Clarity: 3, Insight: 3, Conciseness: 2, Authenticity: 0)
> "[Quote text]"
**Source:** [Attribution]
**Context:** [Context]
**Sub-theme:** [If applicable]
---
[Continue for requested number of quotes]
---
## Quote Breakdown
**By sub-theme:**
- [Sub-theme 1]: [count] quotes
- [Sub-theme 2]: [count] quotes
**By sentiment:**
- Positive: [count] quotes
- Negative/Pain point: [count] quotes
- Neutral: [count] quotes
- Mixed: [count] quotes
**By source:**
- [Transcript 1]: [count] quotes
- [Transcript 2]: [count] quotes
---
## Additional Relevant Quotes (Score 6-7)
[Listing of good but not exceptional quotes, if space allows]
---
## Search Notes
[Any observations about the topic coverage:
- Was topic widely discussed or rare?
- Concentrated in specific transcripts?
- Consistent viewpoint or varied perspectives?
- Related topics that came up frequently?]
```
## Special Handling
### Multiple Perspectives
If topic has varied perspectives:
```markdown
**Perspective A: [Viewpoint]**
"[Quote representing this view]"
**Perspective B: [Contrasting viewpoint]**
"[Quote representing alternative view]"
```
### Sensitive Topics
If quotes contain:
- Negative feedback about specific people/companies
- Confidential information
- Potentially problematic content
Flag these and ask user how to handle:
```
⚠️ Note: Some quotes contain [sensitive content type].
Would you like me to:
1. Anonymize/redact specific names
2. Include with warning
3. Exclude these quotes
```
### Quote Length
For very long relevant passages:
- Extract most impactful 2-3 sentences
- Note that full context available in [source transcript:line]
- Or provide "short version" and "full version"
## Context Preservation
Always provide enough context that:
- Reader understands what participant is responding to
- Quote makes sense without reading full transcript
- Participant's intent is clear
Use bracketed additions if helpful:
> "It [the onboarding process] was confusing and took way too long."
Begin quote selection now.

127
agents/theme-extractor.md Normal file
View File

@@ -0,0 +1,127 @@
---
name: theme-extractor
description: Analyzes interview transcripts to identify recurring themes and patterns
model: sonnet
---
# Theme Extractor Agent
You are a qualitative research analyst who identifies recurring themes and patterns in interview transcripts.
## Your Task
Analyze your assigned interview transcripts and identify 3-5 major themes with supporting evidence.
## Analysis Methodology
### Step 1: Read All Assigned Transcripts
For each transcript:
- Read completely without rushing to conclusions
- Note key topics and concerns mentioned
- Track emotional indicators (frustration, excitement, confusion)
- Identify specific examples and stories participants share
### Step 2: Identify Patterns
Look for patterns across transcripts:
- **Repeated mentions** - Same topic across multiple interviews
- **Consistent pain points** - Similar problems or frustrations
- **Common workflows** - Shared processes or behaviors
- **Frequent requests** - Features or improvements mentioned multiple times
- **Shared contexts** - Similar situations or triggers
### Step 3: Define Themes
For each theme, provide:
**Theme Name:** Clear, descriptive title (3-6 words)
**Description:** 2-3 sentence explanation of what this theme represents
**Frequency:** How many of your assigned transcripts mentioned this theme
**Representative Quotes:** 2-3 compelling quotes that exemplify this theme
- Include speaker/participant attribution if available
- Provide brief context for each quote
**Sub-themes:** If applicable, note related sub-topics within this theme
**Sentiment:** Overall tone (Positive, Negative/Pain Point, Neutral, Mixed)
## Quality Standards
Your themes should be:
**Specific** - Not too broad ("pricing" not "concerns")
**Evidence-based** - Supported by actual quotes
**Significant** - Appeared in multiple transcripts or was emphasized strongly
**Actionable** - Insight that could drive decisions
## Output Format
```markdown
# Theme Analysis Results
**Transcripts analyzed:** [Number] transcripts
**Themes identified:** [Number]
---
## Theme 1: [Theme Name]
**Frequency:** [X] of [Y] transcripts ([percentage]%)
**Sentiment:** [Positive/Negative/Neutral/Mixed]
**Description:**
[2-3 sentence explanation of this theme]
**Sub-themes:**
- [Sub-theme 1]
- [Sub-theme 2]
**Representative Quotes:**
1. > "[Quote 1]"
> — [Participant/Transcript ID]
>
> Context: [When/why this was mentioned]
2. > "[Quote 2]"
> — [Participant/Transcript ID]
3. > "[Quote 3]"
> — [Participant/Transcript ID]
**Why this matters:**
[What this theme suggests about user needs/pain points/opportunities]
---
## Theme 2: [Theme Name]
[Same format...]
---
[Continue for all themes]
---
## Additional Observations
[Any cross-cutting patterns, contradictions, or surprising findings worth noting]
```
## Special Instructions
**If provided with context document (ICP, research goals):**
- Flag themes that align with research objectives
- Note which themes are most relevant to target customer profile
- Indicate unexpected findings that contradict assumptions
**Handle edge cases:**
- If participant is unclear, note ambiguity
- If theme appears in only 1 transcript but is significant, include it but flag as "Single occurrence, high impact"
- If multiple interpretations possible, note alternative readings
Begin your analysis now.

364
commands/analyze-interviews.md Normal file
View File

@@ -0,0 +1,364 @@
---
description: Extract themes and insights from interview transcripts with intelligent agent scaling
argument-hint: [output-name] [transcripts-directory] [--context-file]
---
# Analyze Interview Transcripts
Analyze all interview transcripts in a directory, extract themes, rank insights, and generate a comprehensive report.
## Step 1: Discover and Validate Transcripts
**Scan the transcripts directory ($2):**
- Find all text files (.txt, .md, .doc, .docx)
- Count total transcripts
- Calculate total file size and estimated token count
- Identify any context files (ICP, research goals, methodology docs)
**Look for context file ($3):**
- If --context-file argument provided, read that file
- If not provided, look for common files in directory:
- `icp.md` or `ideal-customer-profile.md`
- `research-goals.md` or `objectives.md`
- `methodology.md`
- `customer-profile.md` or `persona.md`
**Report to user:**
```markdown
## 📁 Transcripts Discovered
**Directory:** $2
**Transcripts found:** [number] files
**Total size:** [size in MB]
**Estimated tokens:** ~[number]k tokens
**Files:**
- [filename 1] ([size])
- [filename 2] ([size])
- [etc.]
**Context documents found:**
- [ICP/research goals/methodology if found, or "None"]
Would you like to proceed with analysis?
```
**Wait for user confirmation.**
---
## Step 2: Determine Agent Scaling Strategy
Based on workload, intelligently determine how many parallel agents to use:
**Agent Scaling Logic:**
```
If 1-3 transcripts:
→ Use 1 agent (sequential analysis is fine)
If 4-8 transcripts:
→ Use 3 parallel agents
→ Divide transcripts evenly (e.g., Agent 1: transcripts 1-3, Agent 2: 4-6, Agent 3: 7-8)
If 9-15 transcripts:
→ Use 5 parallel agents
→ Divide transcripts evenly
If 16+ transcripts:
→ Use 8 parallel agents (optimal for most systems)
→ Divide transcripts evenly
ALSO consider total token count:
If total tokens > 150k:
→ Increase agent count by 1-2
→ Example: 6 transcripts but 200k tokens → use 5 agents instead of 3
If total tokens < 30k:
→ Decrease agent count
→ Example: 10 short transcripts with 25k tokens → use 3 agents instead of 5
```
**Present strategy to user:**
```markdown
## 🤖 Agent Strategy
Based on [number] transcripts totaling ~[number]k tokens, I recommend using **[number] parallel agents** for efficient analysis.
**How it works:**
- Each agent will analyze [number] transcripts
- Agents work simultaneously (faster than sequential)
- All findings will be synthesized into final report
**Agent assignments:**
- Agent 1: [transcript names]
- Agent 2: [transcript names]
- [etc.]
**Estimated time:** [minutes] minutes
Would you like to:
1. ✅ Proceed with this strategy
2. 🔄 Adjust agent count (tell me your preference)
3. ❌ Cancel
```
**Wait for user approval or adjustment.**
---
## Step 3: Launch Theme Extraction Agents
For each agent, use the Task tool with the **theme-extractor** agent:
**Provide each agent with:**
- Its assigned transcripts (full text)
- Context file contents (if available)
- Instructions to identify 3-5 major themes
- Request supporting quotes for each theme
**Track progress:**
```markdown
## 🔍 Extracting Themes
- ✅ Agent 1: Complete (found 4 themes)
- ✅ Agent 2: Complete (found 5 themes)
- 🔄 Agent 3: In progress
- ⏳ Agent 4: Pending
- ⏳ Agent 5: Pending
```
**Once all agents complete, collect results:**
- All themes from all agents
- Supporting quotes
- Frequency counts (how many transcripts mentioned each theme)
---
## Step 4: Synthesize and Rank Themes
Use the **insight-ranker** agent to:
**Consolidate themes:**
- Merge duplicate/similar themes across agents
- Example: "Pricing concerns" + "Cost barriers" → "Pricing and cost concerns"
**Rank themes by:**
1. **Frequency** - How many transcripts mentioned it
2. **Relevance** - Alignment with ICP/research goals (if context provided)
3. **Impact** - Magnitude of the insight (pain point severity, opportunity size)
**Present ranked themes to user:**
```markdown
## 🎯 Top Themes Identified
### Theme 1: [Theme Name]
**Mentioned in:** [X] of [Y] transcripts ([percentage]%)
**Relevance:** [High/Medium/Low based on ICP]
**Key insight:** [One sentence summary]
**Representative quote:**
> "[Quote from transcript]"
> — Participant [number/name]
---
### Theme 2: [Theme Name]
[Same format...]
---
[Continue for top 5-7 themes]
---
**Would you like me to:**
1. ✅ Proceed with final report generation
2. 🔎 Deep dive into specific themes (tell me which ones)
3. 🔄 Re-rank themes with different criteria
```
**Wait for user feedback.**
---
## Step 5: Generate Comprehensive Report
Create a structured markdown report with:
### Report Structure
```markdown
# Interview Analysis Report: [Output Name]
**Analysis date:** [Date]
**Transcripts analyzed:** [Number]
**Context:** [ICP/research goals summary if available]
---
## Executive Summary
[2-3 paragraph overview of most important findings]
---
## Methodology
**Transcripts:**
- [List of transcripts with dates if available]
**Analysis approach:**
- [Number] parallel agents
- Theme extraction and consolidation
- Ranking by frequency and relevance
**Context documents:**
- [List context files if used]
---
## Key Themes
### 1. [Theme Name] ([frequency]% of interviews)
**Summary:** [Detailed explanation of this theme]
**Why it matters:** [Relevance to ICP/business goals]
**Supporting quotes:**
1. "[Quote]" — Participant [ID]
2. "[Quote]" — Participant [ID]
3. "[Quote]" — Participant [ID]
**Recommendations:**
- [Actionable insight 1]
- [Actionable insight 2]
---
[Repeat for each major theme]
---
## Cross-Cutting Insights
[Patterns that span multiple themes]
---
## Recommendations
1. **[Priority]** [Recommendation based on findings]
2. **[Priority]** [Recommendation based on findings]
[etc.]
---
## Appendix: All Themes
[Comprehensive list of all themes found, including lower-frequency ones]
---
**Analysis performed by:** Interview Transcript Analyzer plugin for Claude Code
**Generated:** [Timestamp]
```
**Save report to:** `analysis/$1-analysis.md`
---
## Step 6: Deliver Results
**Report to user:**
```markdown
## ✅ Analysis Complete!
**Report location:** analysis/$1-analysis.md
**Transcripts analyzed:** [number]
**Themes identified:** [number] major themes, [number] total
**Top insights:** [number]
**Report sections:**
- Executive Summary
- Methodology
- [Number] Key Themes (with quotes and recommendations)
- Cross-cutting insights
- Actionable recommendations
**Next steps:**
1. Review the full report
2. Use `/extract-quotes [theme]` to find more quotes on specific themes
3. Use `/rank-insights` to re-rank with different criteria
```
---
## Error Handling
**If no transcripts found:**
```
❌ No transcripts found in directory: $2
Please ensure:
- Directory path is correct
- Transcripts are in supported formats (.txt, .md, .doc, .docx)
- Files are readable
Expected structure:
interviews/
├── interview-1.txt
├── interview-2.txt
└── context/
└── icp.md
```
**If context file specified but not found:**
```
⚠️ Warning: Context file not found: $3
Would you like to:
1. Proceed without context (insights won't be ranked by ICP relevance)
2. Specify a different context file
3. Cancel and add context file first
```
**If agent fails during analysis:**
- Save partial results from completed agents
- Report which agent failed and why
- Offer to retry with fewer agents or different allocation
---
## Usage Examples
```bash
# Basic analysis
/analyze-interviews customer-feedback ./interviews/
# With ICP context for relevance ranking
/analyze-interviews product-research ./research/transcripts/ --context-file ./research/icp.md
# Large-scale analysis (automatically uses more agents)
/analyze-interviews enterprise-study ./enterprise-interviews/
```
---
## Tips for Best Results
**Organize transcripts clearly:**
- Use descriptive filenames (`interview-customer-name.txt`)
- Include dates if relevant (`2024-01-15-interview-acme-corp.txt`)
- Keep transcripts in dedicated directory
**Provide context:**
- ICP document helps rank insights by relevance
- Research goals ensure findings align with objectives
- Methodology doc helps interpret findings appropriately
**For large datasets:**
- The plugin will automatically scale agents (no manual config needed)
- More transcripts = more agents = faster analysis
- Expected speed: ~15-20 transcripts in 15-20 minutes with 8 agents

168
commands/extract-quotes.md Normal file
View File

@@ -0,0 +1,168 @@
---
description: Find representative quotes on specific topics from interview transcripts
argument-hint: [topic] [transcripts-directory] [--max-quotes]
---
# Extract Quotes by Topic
Find the most representative and compelling quotes about a specific topic from interview transcripts.
## Step 1: Scan Transcripts
**Search directory $2 for transcripts:**
- Identify all transcript files
- Count total transcripts available
**Report to user:**
```markdown
## 📁 Transcripts Found
**Directory:** $2
**Transcripts:** [number] files
**Topic to search:** "$1"
Proceeding to extract relevant quotes...
```
---
## Step 2: Extract Relevant Quotes
Use the **quote-selector** agent to:
**For each transcript:**
1. Search for mentions of topic "$1" (and related terms)
2. Extract full quote with context
3. Rate quote quality (clarity, insight value, brevity)
4. Include speaker/participant attribution if available
**Quote selection criteria:**
- **Relevant** - Directly addresses the topic
- **Insightful** - Reveals something meaningful
- **Clear** - Standalone understandable
- **Concise** - Not overly long (prefer 1-3 sentences)
- **Authentic** - Conversational tone preserved
---
## Step 3: Rank and Present Quotes
**Default: Show top 10 quotes**
**If --max-quotes specified: Show that many**
**Present to user:**
```markdown
## 💬 Quotes About: "$1"
Found [total number] relevant quotes across [number] transcripts.
Showing top [$3 or 10] by quality score.
---
### Quote 1 (Score: [X]/10)
> "[Full quote text]"
**Source:** [Transcript name / Participant ID]
**Context:** [Brief context if helpful - when/why this was said]
---
### Quote 2 (Score: [X]/10)
[Same format...]
---
## Quote Categories
**By theme:**
- [Sub-theme 1]: [count] quotes
- [Sub-theme 2]: [count] quotes
**By sentiment:**
- Positive: [count]
- Negative/Pain point: [count]
- Neutral: [count]
---
## Usage Tips
**To find more quotes:**
```bash
/extract-quotes "[topic]" [directory] --max-quotes 20
```
**To search related terms:**
```bash
/extract-quotes "pricing or cost or budget" [directory]
```
```
---
## Step 4: Optional - Export Quotes
**Ask user:**
```
Would you like to:
1. Save these quotes to a file (quotes/$1-quotes.md)
2. Copy to clipboard for pasting elsewhere
3. Just view them here
```
**If user chooses save:**
- Create `quotes/` directory if needed
- Save as `quotes/$1-quotes.md` with all quotes in markdown format
---
## Error Handling
**If topic not found in any transcript:**
```
⚠️ No quotes found for topic: "$1"
Suggestions:
- Try broader search terms
- Check transcript file names to ensure they're included
- Try related terms: /extract-quotes "[alternative terms]"
```
**If no transcripts in directory:**
```
❌ No transcripts found in: $2
Please ensure:
- Directory path is correct
- Transcripts are in .txt or .md format
```
---
## Usage Examples
```bash
# Find quotes about pricing
/extract-quotes pricing ./interviews/
# Find more quotes with custom limit
/extract-quotes "product feedback" ./interviews/ --max-quotes 20
# Search multiple related terms
/extract-quotes "onboarding or getting started" ./customer-calls/
```
---
## Tips
**Effective search terms:**
- Single concepts: `pricing`, `features`, `support`
- Multi-word phrases: `"user experience"`, `"biggest challenge"`
- Related terms with OR: `"price or cost or budget"`
**Quote quality:**
- Scores 8-10: Exceptional quotes (use in presentations)
- Scores 6-7: Good quotes (use in reports)
- Scores 4-5: Useful context (reference material)

252
commands/rank-insights.md Normal file
View File

@@ -0,0 +1,252 @@
---
description: Re-rank insights from previous analysis with different criteria
argument-hint: [analysis-file] [--criteria]
---
# Re-Rank Insights
Re-rank insights from a previous analysis using different criteria (frequency, impact, relevance, or custom).
## Step 1: Load Previous Analysis
**Read the analysis file ($1):**
- Parse all themes/insights from the file
- Extract frequency data
- Note any existing rankings
**Report to user:**
```markdown
## 📊 Analysis Loaded
**File:** $1
**Themes found:** [number]
**Current ranking:** [Current criteria if specified]
**Available themes:**
1. [Theme name] - Currently ranked #1
2. [Theme name] - Currently ranked #2
[etc.]
How would you like to re-rank these insights?
```
---
## Step 2: Select Ranking Criteria
**Present options to user:**
```markdown
## 🎯 Ranking Criteria Options
Choose how to rank insights:
1. **Frequency** - Most mentioned themes first
- Best for: Understanding what came up most often
- Data available: ✅
2. **Impact** - Highest impact/severity first
- Best for: Prioritizing what to act on
- Requires: Manual input or impact assessment
3. **Relevance to ICP** - Best fit with ideal customer profile
- Best for: Focusing on target audience needs
- Requires: ICP document for comparison
4. **Business Value** - Revenue/strategic opportunity size
- Best for: ROI-focused prioritization
- Requires: Business context assessment
5. **Ease of Implementation** - Quick wins first
- Best for: Tactical planning
- Requires: Feasibility assessment
6. **Custom** - Provide your own criteria
- Example: "Rank by technical complexity" or "Rank by customer segment"
Which criteria would you like to use? [1-6 or describe custom]
```
**Wait for user selection.**
---
## Step 3: Apply Ranking Criteria
Based on user's choice:
### If Frequency (Option 1):
- Simply re-sort themes by mention count
- No additional analysis needed
### If Impact (Option 2):
Use the **insight-ranker** agent to:
- Review each theme
- Assess potential impact (pain point severity, opportunity size)
- Rank accordingly
### If Relevance to ICP (Option 3):
**Ask for ICP document:**
```
Please provide path to your ICP document, or I'll look for:
- icp.md
- ideal-customer-profile.md
- customer-profile.md
in the analysis directory
```
Then use **insight-ranker** agent to:
- Read ICP document
- Score each theme by alignment with ICP characteristics
- Rank by relevance score
### If Business Value (Option 4):
**Prompt user for context:**
```
To rank by business value, I need to understand:
1. Revenue impact indicators (which themes suggest upsell/retention opportunities?)
2. Strategic priorities (which align with company goals?)
Please provide:
- Business goals document, OR
- Quick description of strategic priorities
Or I can make best-effort assessment based on theme content.
```
### If Custom (Option 6):
**Ask user to clarify:**
```
Please describe your custom ranking criteria:
Example: "Rank by technical complexity, easiest first"
Example: "Rank by enterprise vs SMB relevance"
Example: "Rank by urgency based on competitive pressure"
```
Then use **insight-ranker** agent with custom instructions.
---
## Step 4: Present Re-Ranked Results
```markdown
## ✅ Insights Re-Ranked: [Criteria]
### New Ranking:
#### #1: [Theme Name]
**Previous rank:** #[old position]
**Score:** [Score based on criteria] / 10
**Why this ranking:** [Brief explanation of why it ranks here]
**Key insight:** [One sentence summary]
---
#### #2: [Theme Name]
[Same format...]
---
[Continue for all themes]
---
## Ranking Changes
**Moved up:**
- [Theme]: #[old] → #[new] (↑[change])
**Moved down:**
- [Theme]: #[old] → #[new] (↓[change])
**Stayed same:**
- [Theme]: #[position]
---
## Next Steps
Would you like to:
1. Save this new ranking to file (updates original or creates new)
2. Compare rankings side-by-side (original vs new)
3. Try different criteria
4. Export top 3 insights as summary
```
---
## Step 5: Save or Export (Optional)
**If user chooses to save:**
```markdown
Options:
1. ✅ Update original file ($1) with new ranking
2. 📄 Save as new file (analysis/$1-reranked-[criteria].md)
3. 📋 Just show me the ranking (don't save)
Which would you prefer?
```
---
## Error Handling
**If analysis file doesn't exist:**
```
❌ Analysis file not found: $1
Available analysis files in current directory:
- [file1]
- [file2]
Usage: /rank-insights [analysis-file] [--criteria]
```
**If insufficient data for chosen criteria:**
```
⚠️ Not enough data to rank by "[criteria]"
This ranking method requires:
- [What's needed]
Would you like to:
1. Choose different criteria
2. Provide required context
3. Cancel
```
---
## Usage Examples
```bash
# Rank by frequency (simplest)
/rank-insights analysis/customer-research-analysis.md --criteria frequency
# Rank by ICP relevance
/rank-insights analysis/interviews-analysis.md --criteria relevance
# Custom ranking
/rank-insights analysis/feedback-analysis.md --criteria "urgency based on churn risk"
```
---
## Tips for Effective Ranking
**Frequency ranking:**
- Good for: Identifying consensus themes
- Watch out for: Silent but critical minority opinions
**Impact ranking:**
- Good for: Prioritizing action items
- Consider: Both pain severity and opportunity size
**Relevance ranking:**
- Good for: Focusing on target audience
- Requires: Well-defined ICP
**Custom ranking:**
- Most flexible
- Best when you have specific strategic context
- Examples: "competitive urgency", "implementation cost", "customer segment"

65
plugin.lock.json Normal file
View File

@@ -0,0 +1,65 @@
{
"$schema": "internal://schemas/plugin.lock.v1.json",
"pluginId": "gh:animalzinc/claude-plugins:plugins/interview-transcript-analyzer",
"normalized": {
"repo": null,
"ref": "refs/tags/v20251128.0",
"commit": "7703523e37172270e7319dd94d33a101e3f4c6d6",
"treeHash": "f40a8408dfc0dcd00b6ff6be897736637ed4bc09035c3246119491c17d93e0ef",
"generatedAt": "2025-11-28T10:13:44.524241Z",
"toolVersion": "publish_plugins.py@0.2.0"
},
"origin": {
"remote": "git@github.com:zhongweili/42plugin-data.git",
"branch": "master",
"commit": "aa1497ed0949fd50e99e70d6324a29c5b34f9390",
"repoRoot": "/Users/zhongweili/projects/openmind/42plugin-data"
},
"manifest": {
"name": "interview-transcript-analyzer",
"description": "Analyze interview transcripts with unlimited token limits. Extract themes, rank insights by relevance, and find representative quotes. Intelligently scales agents based on workload.",
"version": "1.0.0"
},
"content": {
"files": [
{
"path": "README.md",
"sha256": "4f9386a10e6994f0dcb8fa03705251844557dd2d976d5170cc2dd5eff2147359"
},
{
"path": "agents/quote-selector.md",
"sha256": "81d555d4e926e231baf2d9190fbc51505a60a89d9e6da199a78bd17f0c6472ad"
},
{
"path": "agents/theme-extractor.md",
"sha256": "e3bc03e93c325468ea99583405f2b1616cefdf86b658074c387fa75555f43086"
},
{
"path": "agents/insight-ranker.md",
"sha256": "2f0f9c764c60ddd77a31034285503748b32fcdedf981cec8ea255d302a015b85"
},
{
"path": ".claude-plugin/plugin.json",
"sha256": "906e9e970c930d3f209b503be81c61191b3feab4ee3562d0bf20403b6daa905c"
},
{
"path": "commands/rank-insights.md",
"sha256": "f96735a66e27703617eda592884be001bbd2c3adf67188c1653d9c552ef1b59d"
},
{
"path": "commands/extract-quotes.md",
"sha256": "d51015219fc869c67a17863a6a9bb11d31cf02c291a976c7ffc364f7f6beafb7"
},
{
"path": "commands/analyze-interviews.md",
"sha256": "1579eca15ed6200a1256c155d50764eab0a9cec99b548839b1f40a0e3ac93532"
}
],
"dirSha256": "f40a8408dfc0dcd00b6ff6be897736637ed4bc09035c3246119491c17d93e0ef"
},
"security": {
"scannedAt": null,
"scannerVersion": null,
"flags": []
}
}