name: clifton
description: CliftonStrengths analysis with individual coaching, team comparison, and multi-format output (markdown, HTML, slides)
agent: advisor
version: 1.0
classification: public
last_updated: 2026-01-26
effort_default: STANDARD

⛔ DUAL-PATH ROUTING - READ THIS FIRST

STOP. This skill requires the advisor agent for complex requests.

Identity check: If you are NOT the advisor agent AND your request is complex
(personalized coaching, blindspot analysis, team comparison) → DELEGATE NOW:

typescript Task(subagent_type="advisor", prompt="Execute clifton skill. Request: {user_request}")

DO NOT proceed if you lack advisor expertise for:
- Personalized strengths coaching
- Blindspot detection and analysis
- Team dynamics assessment
- Development strategy planning

Path 1 - Simple (Tier 1/Haiku): General CliftonStrengths information
- "What are the 34 CliftonStrengths themes?"
- "Explain the Activator strength"
- Routes directly, no delegation needed

Path 2 - Complex (Advisor): Personalized analysis
- "Analyze my CliftonStrengths results"
- "Help me with team comparison"
- Requires advisor delegation

name: clifton
description: CliftonStrengths analysis with individual coaching, team comparison, and multi-format output (markdown, HTML, slides). Brutally honest blindspot analysis.
agent: advisor
version: 1.0
classification: public
last_updated: 2026-01-26
effort_default: STANDARD

FOR AI AGENTS: CliftonStrengths coaching and team analysis.
Load when: user mentions "CliftonStrengths", "strengths analysis", "Gallup", "theme analysis", "blindspots", "team strengths"

CliftonStrengths Analysis Skill

Deep-dive strengths coaching with brutally honest blindspot analysis and multi-format output.

Model Tier Routing

This skill uses the three-tier orchestration engine for cost-optimal execution:

Routing Decision Tree:
- Strengths theme explanations: Tier 1 (Free) - $0.00 (e.g., "Explain Analytical theme")
- Individual coaching and analysis: Tier 2b (Grok 4.1 Fast) - $0.70/1M ← Primary (complex coaching)
- Team comparison and blindspot analysis: Tier 2b (Grok 4.1 Fast) - $0.70/1M (multi-step reasoning)

Why Tier 2b for Clifton?
- Complex multi-step coaching logic
- Team comparison orchestration
- Brutally honest blindspot analysis (multi-phase reasoning)
- Large context for full strengths interpretation
- Cost savings vs. Sonnet: 71% ($0.70 vs $2.40 per 1M)

How it works:
1. You invoke /clifton with your strengths profile
2. Base Claude's tier selector analyzes complexity and coaching depth
3. Simple explanations route to Tier 1 (free)
4. Full coaching analysis routes to Tier 2b for complex reasoning
5. Team analysis uses Tier 2b for multi-person comparison

Cost Model:
- Theme explanation: $0 (Tier 1)
- Individual coaching: ~$0.70 (Tier 2b for deep analysis)
- Team comparison: ~$1.40 (full orchestration)

Reference: See docs/skill-tier-integration.md for complete skill mapping.

Pre-flight Checklist (MANDATORY)

STOP! Before executing this skill:

• [ ] Read this SKILL.md completely
• [ ] Load knowledge base: docs/cliftonstrengths-themes.md
• [ ] Detect mode: Individual vs Team
• [ ] Detect output format: md (default), html, slides

Core Philosophy

Brutally Honest: This is NOT validation. It's honest assessment of strengths AND limitations.

Name It, Claim It, Aim It: Gallup's framework - recognize patterns, accept them, leverage them.

Bottom 5 Matter: Your weaknesses are real. Don't fix them - partner around them.

Critical Rules

1. Load Knowledge Base First - docs/cliftonstrengths-themes.md has all 34 themes
2. Brutally Honest - No sugar-coating blindspots and limitations
3. Bottom 5 = Partner - Never suggest "developing" bottom themes
4. Domain Gaps = Structural - Zero themes in a domain is a real limitation

Resource Auto-Detection

Individual Mode - Scan input/individual/ folder:

skills/clifton/input/
├── individual/           # Single person analysis (default)
│   ├── report.pdf       # Primary report location
│   └── split/           # Split chunks (auto-created if PDF too large)
│       ├── report_chunk_1.pdf
│       └── report_chunk_2.pdf
└── README.md

Team Mode - Scan input/team/ folder:

skills/clifton/input/
└── team/                # Team member reports
    ├── alice.pdf
    ├── bob.pdf
    ├── carol.pdf
    └── split/           # Split chunks for team members (auto-created)
        ├── alice/
        │   ├── alice_chunk_1.pdf
        │   └── alice_chunk_2.pdf
        └── bob/

Detection Logic:
1. Check skills/clifton/input/individual/ for PDF files (default mode)
2. If multiple PDFs found in individual/, prefer report.pdf or ask user
3. If input/team/ folder has PDFs, offer team comparison mode
4. If PDF > 5MB or read fails, auto-trigger PDF splitter workflow
5. If no PDFs found, ask user to provide report path or list themes manually

Supported Report Types:
- Full 34 report (all themes ranked) - Recommended
- Top 5 report (signature themes only - limited blindspot analysis)

Mode Detection

Output Format Options

Default behavior: Generates all formats automatically (Markdown + HTML + Slides)

Override flags (optional):

Slides Options (When format=slides)

Marp-based presentations with theme selection and optional image generation.

Quick examples:

/clifton --format=slides              # Dark theme, no images
/clifton --format=slides --theme=light --images=yes

Complete documentation: docs/slides-and-marp.md

5-Phase Workflow (Individual Mode)

┌─────────┐     ┌──────────┐     ┌─────────┐     ┌──────────┐     ┌─────────┐
│  PARSE  │────▶│ ANALYZE  │────▶│ BLINDSPOT│────▶│ DEVELOP  │────▶│ DELIVER │
│         │     │          │     │         │     │          │     │         │
└────┬────┘     └────┬─────┘     └────┬────┘     └────┬─────┘     └────┬────┘
     │               │                │               │               │
     ▼               ▼                ▼               ▼               ▼
  Extract        Pattern          Bottom 5       Name/Claim      Multi-format
  Themes         Analysis         + Gaps         /Aim Plan       Output

Phase 1: PARSE (Extract Themes)

Extract from CliftonStrengths Report:

1. Theme Rankings - All 34 (or Top 5 if limited)
2. Domain Distribution - Count per domain
3. Signature Themes - Top 5 as core identity

The Four Domains:

Note: Influencing is statistically RAREST (only 15% of Top 5s)

Phase File: phases/01-parse.md

Phase 2: ANALYZE (Pattern Analysis)

Focus on #1 + #2 Theme Interaction

The top two themes create a unique behavioral signature.

Analysis Framework:
- How do these amplify each other?
- What unique behavior does this combination create?
- What's the "signature move"?
- Career/role implications?
- Risk patterns?

Knowledge Base Lookup:
- Check if common pairing (Input+Learner, Woo+Communication)
- Check if rare tension pairing (Deliberative-Woo, Developer-Command)
- Pull combination insights from theme reference

Phase File: phases/02-analyze.md

Phase 3: BLINDSPOT (Honest Limitations)

Most Important Phase - No Sugar-Coating

Bottom 5 Themes (if Full 34)

These are natural non-talents. Gallup research: you CANNOT effectively develop these.

Strategy: Partner with someone strong here. Don't try to fix.

Domain Gaps

Zero themes in a domain = significant structural blindspot.

Overuse Patterns (Top 10)

Every strength has a shadow side (basement vs balcony).

Phase File: phases/03-blindspot.md

Phase 4: DEVELOP (Recommendations)

"Name It, Claim It, Aim It" Framework

Name It

• Recognize your specific patterns
• Understand theme interactions
• Know your situational triggers

Claim It

• Accept these as natural patterns
• Stop fighting your nature
• Embrace what you're NOT good at
• Find partners who complement weaknesses

Aim It

1. Role/Environment Fit - Where do you thrive vs drain?
2. Team Complementary - What themes do you need around you?
3. Action Items - Specific, actionable recommendations

Partnership Recommendations (Bring/Need Framework)

For each Top 5 theme, identify what you bring and what you need from partners.

Phase File: phases/04-develop.md

Phase 5: DELIVER (Multi-Format Output)

Generate deliverables in requested format(s).

Markdown Output (Default)

skills/clifton/output/{Name}-{YYYY-MM-DD}/
├── strengths-analysis.md     # Full theme breakdown
├── blindspot-report.md       # Brutally honest limitations
└── development-plan.md       # Name/Claim/Aim recommendations

HTML Output

Same content, styled HTML with CliftonStrengths domain colors:
- Purple (Executing), Orange (Influencing), Blue (Relationship), Green (Strategic)

skills/clifton/output/{Name}-{YYYY-MM-DD}/
├── strengths-analysis.html
├── blindspot-report.html
└── development-plan.html

Slides Output (Marp)

Generate Marp-compatible markdown with 8-slide structure (title, themes, domain distribution, signature pattern, blindspots, development, etc.)

See: docs/slides-and-marp.md for complete slide structure and Marp conversion commands.

Phase File: phases/05-deliver.md

Output Structure

All deliverables in ONE folder: skills/clifton/output/{Name}-{YYYY-MM-DD}/

Core files (always):
- strengths-analysis.md - Full theme breakdown
- blindspot-report.md - Brutally honest limitations
- development-plan.md - Name/Claim/Aim recommendations

Additional files (format-dependent):
- HTML: *.html versions of core files
- Slides: strengths-slides.md + strengths-slides.html
- Images: images/*.png (if --images=yes)

Optional: readme.md for quick reference (only if helpful)

Team Mode (6 Phases)

For team analysis, add Phase T0 and modify phases:

Team Output (Markdown - default):

output/Team-{Name}-{YYYY-MM-DD}/
├── team-grid.md
├── domain-analysis.md
├── partnership-map.md
└── team-development.md

Additional files by format:
- --format=html: Add .html versions of all files
- --format=slides: Add team-strengths-slides.md + team-strengths-slides.html
- --format=all: All markdown + HTML + slides

Progress Tracking

Use TodoWrite to display progress:

[ ] Phase 1: Parse - Extract themes and domain distribution
[ ] Phase 2: Analyze - Identify dominant pattern and tensions
[ ] Phase 3: Blindspot - Document limitations (brutally honest)
[ ] Phase 4: Develop - Create Name/Claim/Aim recommendations
[ ] Phase 5: Deliver - Generate output in requested format(s)

Marp Slide Templates

Template files: templates/marp-dark-theme.md | templates/marp-light-theme.md

Complete documentation: docs/slides-and-marp.md (includes slide structure, image placement, class directives)

Model Selection

Default: Sonnet (analysis, document generation)
Knowledge Base: Load docs/cliftonstrengths-themes.md at start

Dependencies

Required: PDF Splitter (built-in), Marp CLI (auto-install)
Optional: OpenRouter API (image generation)

Complete documentation: docs/dependencies.md (includes installation, usage, error recovery)

Error Recovery

Skill Resources

Reference Files

• docs/cliftonstrengths-themes.md - All 34 themes knowledge base
• docs/image-style-presets.md - Image generation style presets
• docs/openrouter-image-api.md - API documentation

Template Files

• templates/marp-dark-theme.md - Dark theme CSS
• templates/marp-light-theme.md - Light theme CSS

Scripts

• scripts/pdf-splitter.ts - Split large PDFs into readable chunks
• scripts/generate-slide-image.ts - TypeScript image generator

Related Skills

• /career - Job application (separate, not integrated)
• /mentorship - Learning roadmaps and career development

Version: 1.3
Last Updated: 2026-01-14
Status: Active
Recent Update: Marp template v2.0 - dark navy theme matching HTML reports, improved fonts (0.75em body, readable hierarchy)