feat: Add 5 New Analytics Agents

[Daniel Beer (DanielB945)]

Summary

Adds five new agents to support the full analytics workflow: data validation, event discovery, presentations, release tracking, and user research.

What's Included

1. Data Validation Agent (agents/data-validation/SKILL.md)

Trigger: "Validate data quality", "Check for NULLs/gaps", "Verify schema"

Validates data quality in BigQuery tables and query results before analysis.

What it validates:

• NULL checks: Unexpected NULL values in key columns
• Date gaps: Missing dates in time series data
• Invalid values: Negative counts, invalid percentages, out-of-range values
• Schema validation: Column names and types match expectations
• Row count checks: Unexpectedly low or zero row counts
• Duplicates: Duplicate primary keys
• Business logic: Funnel violations, retention increases, impossible metrics

Key features:
✅ Severity classification (🔴 Critical / 🟡 Warning / 🟢 Info)
✅ Actionable recommendations for each issue
✅ Formatted validation reports with summary
✅ Common validation patterns (time series, user data, aggregated metrics)

2. Context Helper (Events) Agent (agents/context-helper/SKILL.md)

Trigger: "What events does [feature] track?", "Explain event [name]", "Find events"

Helps find and explain events for analytics work by searching event registry, code, and Figma.

What it provides:

• Event discovery: Find all events for a feature or keyword
• Event definitions: Explain what each event tracks and when it fires
• Event context: When/where event is triggered in the product
• Usage examples: SQL patterns for using the event
• Data availability: Which tables contain the event
• Related events: Other events in the same flow

Key features:
✅ Searches shared/event-registry.yaml first
✅ Falls back to GitHub code search and Figma
✅ Provides SQL examples for common use cases
✅ Warns about gotchas (e.g., action_category filter)
✅ Shows related events in feature flows

3. Presentation Agent (agents/presentation/SKILL.md)

Trigger: "Create presentation", "Turn findings into slides", "Format for stakeholders"

Creates presentation decks and slides from analytics findings for stakeholder presentations.

What it creates:

• Executive decks: High-level insights for leadership
• QBR presentations: Quarterly business reviews for enterprise customers
• Feature analysis: Deep-dive presentations on feature performance
• Experiment results: A/B test or feature launch results
• Strategy reviews: Data-driven recommendations for product decisions

Key features:
✅ Standard presentation structures (Executive, Feature, Experiment)
✅ One message per slide with clear titles
✅ Visual-first design (charts take center stage)
✅ Executive summary with key takeaways
✅ Actionable recommendations with owners/timelines

4. Release Agent (agents/release/SKILL.md)

Trigger: "Track [feature] release", "Analyze launch performance", "Monitor adoption"

Tracks and analyzes feature releases and product launches. Monitors adoption, performance, and impact.

What it tracks:

• Adoption metrics: Feature DAU, penetration rate, cumulative adopters
• Usage patterns: Frequency, session depth, advanced usage
• User segments: Which tiers/cohorts adopt fastest
• Performance vs goals: Actual vs projected adoption
• Impact assessment: Retention lift, engagement lift, revenue impact
• Comparison: Performance vs other releases

Key features:
✅ Adoption curves and cohort analysis
✅ Success metrics defined upfront
✅ Status indicators (🟢 On Track / 🟡 Monitor / 🔴 Below Target)
✅ Retention impact comparison (feature users vs non-users)
✅ Release report templates

5. User Research Agent (agents/user-research/SKILL.md)

Trigger: "Segment users", "Create personas", "User journey analysis", "Find power users"

Conducts quantitative user research using behavioral data. Identifies segments, personas, and opportunities.

What it provides:

• User segmentation: Behavioral clusters and personas
• User journeys: Common paths through the product
• Pain points: Where users struggle or drop off
• Power user patterns: What high-value users do differently
• Churn signals: Behaviors that predict churn
• Opportunity identification: Unmet needs or underserved segments

Key features:
✅ Behavioral segmentation SQL patterns
✅ Persona templates with metrics
✅ Journey analysis (entry points, drop-offs, success paths)
✅ Power user identification and analysis
✅ Churn prediction and risk assessment

Updated Files

CLAUDE.md

Added routing for all 5 new agents with clear trigger phrases.

README.md

• Updated repo structure to list all 5 agents
• Added detailed descriptions in "Current Agents" section
• Added to agent summary table

Folder Structure

agents/
  ├── data-validation/SKILL.md
  ├── context-helper/SKILL.md
  ├── presentation/SKILL.md
  ├── release/SKILL.md
  └── user-research/SKILL.md

Key Design Principles

All agents follow these principles:
✅ Reference shared files - No duplication (rule #15 compliant)
✅ Clear workflows - Step-by-step instructions
✅ Actionable outputs - Not just data, but recommendations
✅ Multiple use cases - Flexible for different needs
✅ SQL examples - Concrete patterns for common analyses

Use Cases

Data Validation:

• Before presenting dashboard results
• After major data pipeline changes
• When debugging unexpected metric values

Context Helper (Events):

• Starting new feature analysis
• Understanding event tracking
• Finding events for specific use case

Presentation:

• Board meetings and QBRs
• Feature launch reviews
• Experiment result presentations

Release:

• New feature launch tracking
• Adoption monitoring
• Release performance comparisons

User Research:

• Understanding user base
• Creating marketing personas
• Identifying growth opportunities
• Predicting churn

Rules Applied

✅ Rule #15 compliance - All agents reference shared files (no duplication)
✅ Rule #4 compliance - Always read relevant shared files before work
✅ Consistent structure - All agents follow similar SKILL.md format

Total additions: 1,627 lines across 7 files 🚀

🤖 Generated with Claude Code

[Assaf Hay Eden (AssafHayEden)]

add all shared knoweldge files