@rfxlamia/skillkit 1.0.0 → 1.2.0

package/skills/skills/creative-copywriting/references/hook-anatomy.md

@@ -0,0 +1,169 @@
+# Hook Anatomy: The Science of the First 3 Seconds
+
+## Why Hooks Matter
+
+Your hook is your content's first impression. In 2025's attention economy:
+- Users decide to engage in **0.3-3 seconds**
+- 47% of content is scrolled past without engagement
+- The algorithm measures **stop rate** as a key signal
+
+## The Hook Formula
+
+```text
+HOOK = [Emotion Trigger] + [Specificity] + [Open Loop]
+```
+
+### Component 1: Emotion Trigger
+
+The emotional payload that stops the scroll.
+
+**High-Performance Emotions:**
+| Emotion | Stop Rate | Best Format |
+|---------|-----------|-------------|
+| Surprise | 12.3% | Pattern interrupt |
+| Curiosity | 9.8% | Incomplete info |
+| Fear | 8.7% | Warning/consequence |
+| Desire | 8.1% | Transformation promise |
+| Belonging | 7.4% | Identity statements |
+
+### Component 2: Specificity
+
+Vague claims = ignored. Specific details = credibility.
+
+**Transform:**
+- ❌ "I grew my account fast"
+- ✅ "I grew from 0 to 47,000 followers in 63 days"
+
+**Specificity Types:**
+- Numbers: 47.3% (not "almost half")
+- Time frames: 63 days (not "a few months")
+- Money: $10,847 (not "five figures")
+- Quantities: 1,247 posts analyzed (not "thousands")
+
+### Component 3: Open Loop
+
+The unfinished thought that demands closure.
+
+**Open Loop Techniques:**
+1. **Incomplete sentence**: "The one thing that changed everything was..."
+2. **Unanswered question**: "What if everything you know is wrong?"
+3. **Story gap**: "Last week something happened that I never expected..."
+4. **Promise unfulfilled**: "Here's the secret nobody talks about..."
+
+## Hook Types Deep Dive
+
+### Visual Hooks (2025 Trending)
+
+Movement and visual contrast now outperform verbal hooks.
+
+**Techniques:**
+- Drop an object (pattern interrupt)
+- Sudden zoom/camera movement
+- Unexpected location
+- Stop-motion effect
+- Face close-up with emotion
+
+**Why it works:** 50% of users watch without sound. Visual hooks work regardless of audio.
+
+### Curiosity Gap Hooks
+
+Create an information vacuum that demands filling.
+
+**Formula:** `[Partial information] + [Implied value]`
+
+**Examples:**
+- "I discovered the one thing that..." (what thing?)
+- "After 5 years, I finally know..." (know what?)
+- "The truth about [topic] that..." (what truth?)
+
+### Controversy Hooks
+
+Challenge existing beliefs to trigger engagement.
+
+**Formula:** `[Common belief] + [Contradiction]`
+
+**Examples:**
+- "Everyone says post daily. They're wrong."
+- "Followers don't matter. Here's why."
+- "The algorithm isn't broken. You are."
+
+**Warning:** Must deliver evidence. Pure controversy without substance backfires.
+
+### Personal Story Hooks
+
+Vulnerability creates instant connection.
+
+**Formula:** `[Failure/struggle] + [Specific detail]`
+
+**Examples:**
+- "I lost 10,000 followers in 7 days doing this..."
+- "My biggest mistake cost me $47,000..."
+- "I almost quit last month. Here's what changed..."
+
+## Platform-Specific Optimization
+
+### Instagram Carousels
+- Hook on **Slide 1 AND Slide 2** (algorithm resurfaces)
+- Text: Large, readable in 2 seconds
+- Visual hook > Verbal hook
+- First 15 characters critical for search
+
+### X/Twitter
+- First 50 characters = hook zone
+- Strategic truncation (let algorithm cut at curiosity point)
+- Question or incomplete list forces expansion
+- Thread emoji (🧵) signals value ahead
+
+### Reels/TikTok
+- First 1 second = visual pattern interrupt
+- Text hook + verbal hook (sound on/off coverage)
+- Face > No face (1.3x engagement)
+- Movement in first frame
+
+## Hook Testing Framework
+
+### A/B Test Variables
+1. Emotion type (fear vs curiosity vs desire)
+2. Specificity level (vague vs detailed)
+3. Format (question vs statement vs visual)
+4. Length (short punch vs detailed setup)
+
+### Metrics to Track
+- Stop rate (reached / impressions)
+- Swipe-through rate (carousels)
+- Expand rate (X/Twitter)
+- Watch time (video first 3 sec)
+
+### Iteration Process
+1. Create 3 variations (different emotions)
+2. Test with same content body
+3. Measure stop/engagement rate
+4. Double down on winner
+5. Repeat
+
+## Common Hook Mistakes
+
+### Mistake 1: Clickbait Without Payoff
+- Hook promises something content doesn't deliver
+- Result: High initial engagement, poor retention, damaged trust
+
+### Mistake 2: Too Many Power Words
+- "SHOCKING SECRET REVEALED: ULTIMATE GUIDE"
+- Result: Manipulation feel, audience skepticism
+
+### Mistake 3: Generic Hooks
+- "Want to grow? Read this."
+- Result: No emotional trigger, easily scrolled past
+
+### Mistake 4: Platform Mismatch
+- Long text hook on visual-first platform
+- Result: Ignored by visual-scanning users
+
+## Quick Reference: Hook Templates
+
+**Curiosity:** "I discovered the one thing that [outcome]..."
+**Controversy:** "Everyone says [belief]. They're wrong."
+**Personal:** "I [failure/struggle] and here's what I learned..."
+**Data:** "[Surprising stat]. Here's what it means..."
+**Question:** "What if [challenging assumption]?"
+**Visual:** *[Action]* + "This is [metaphor for concept]"

package/skills/skills/creative-copywriting/references/power-word-science.md

@@ -0,0 +1,134 @@
+# Power Word Science: Psychology Behind Trigger Words
+
+## Table of Contents
+- [What Are Power Words](#what-are-power-words)
+- [The Neuroscience](#the-neuroscience)
+- [Category Deep Dives](#category-deep-dives)
+- [Strategic Usage](#strategic-usage)
+- [Common Mistakes](#common-mistakes)
+
+## What Are Power Words
+
+Power words are specific terms that trigger emotional responses, bypassing logical processing and activating the limbic system (emotional brain) directly.
+
+**Key Characteristics:**
+- Evoke immediate emotional reaction
+- Increase engagement metrics measurably
+- Work across cultures (with localization)
+- Can backfire if overused
+
+## The Neuroscience
+
+### Dual Processing Theory
+
+**System 1 (Fast/Emotional):**
+- Processes power words instantly
+- Triggers emotional response
+- No conscious analysis
+- Where hooks live
+
+**System 2 (Slow/Logical):**
+- Processes factual claims
+- Requires mental effort
+- Easily fatigued
+- Where proof lives
+
+**Implication:** Hook with System 1 (emotion), support with System 2 (logic).
+
+### Emotional Categories
+
+| Emotion | Brain Region | Trigger Words |
+|---------|--------------|---------------|
+| Fear | Amygdala | Warning, Danger, Risk |
+| Desire | Nucleus Accumbens | Ultimate, Dream, Transform |
+| Trust | Prefrontal Cortex | Proven, Guaranteed, Authentic |
+| Curiosity | Dopamine System | Secret, Discover, Hidden |
+
+## Category Deep Dives
+
+### Urgency Words
+**Psychology:** Loss aversion—we fear losing more than we enjoy gaining.
+
+**Top Performers:** Now, Limited, Ending, Final, Before
+
+**Usage:** Time-sensitive CTAs, launches, limited offers
+
+**Caution:** Fake urgency destroys trust
+
+### Exclusivity Words
+**Psychology:** Scarcity + Status—exclusive things feel more valuable.
+
+**Top Performers:** Secret, Insider, Hidden, Private, Exclusive
+
+**Usage:** Premium content, insider knowledge, community building
+
+**Caution:** Overuse dilutes exclusivity
+
+### Curiosity Words
+**Psychology:** Information gap theory—brain needs closure.
+
+**Top Performers:** Discover, Reveal, Truth, Finally, Uncover
+
+**Usage:** Hooks, headlines, open loops
+
+**Caution:** Must deliver payoff
+
+### Fear Words
+**Psychology:** Survival instinct—we prioritize threats.
+
+**Top Performers:** Warning, Mistake, Avoid, Risk, Danger
+
+**Usage:** Problem-aware content, cautionary tales
+
+**Caution:** Always offer solution after fear
+
+### Desire Words
+**Psychology:** Aspirational identity—who we want to become.
+
+**Top Performers:** Transform, Ultimate, Perfect, Dream, Unlock
+
+**Usage:** Transformation content, aspirational marketing
+
+**Caution:** Don't overpromise
+
+## Strategic Usage
+
+### The 1-2 Rule
+Maximum 1-2 power words per hook.
+
+**Why:**
+- More = manipulation feel
+- Audience becomes desensitized
+- Credibility decreases
+
+**Good:** "The secret strategy that transformed my reach"
+**Bad:** "The SHOCKING SECRET strategy that will INSTANTLY TRANSFORM your reach FOREVER"
+
+### Emotion-Content Matching
+
+| Content Type | Primary Emotion | Power Words |
+|--------------|-----------------|-------------|
+| Educational | Curiosity | Discover, Reveal, Truth |
+| Promotional | Urgency + Desire | Limited, Transform, Now |
+| Story | Trust + Vulnerability | Real, Authentic, Honest |
+| Warning | Fear + Solution | Avoid, Instead, Better |
+
+### Placement Optimization
+
+**Hook Position:** First 15 characters
+**CTA Position:** Near action verb
+**Avoid:** Mid-content (breaks flow)
+
+## Common Mistakes
+
+### Mistake 1: Power Word Stuffing
+Too many triggers = spam feel
+
+### Mistake 2: Emotion Mismatch
+Urgency words on evergreen content confuses audience
+
+### Mistake 3: Empty Promises
+"Secret" without actual exclusive info = clickbait
+
+### Mistake 4: Fear Without Solution
+Creating anxiety without relief damages relationship

package/skills/skills/creative-copywriting/references/storytelling-frameworks.md

@@ -0,0 +1,157 @@
+# Storytelling Frameworks: Complete Narrative Structures
+
+## Table of Contents
+- [Why Stories Work](#why-stories-work)
+- [Core Frameworks](#core-frameworks)
+- [Social Media Adaptations](#social-media-adaptations)
+- [Framework Selection Guide](#framework-selection-guide)
+
+## Why Stories Work
+
+### Neuroscience of Story
+
+**Neural Coupling:** When hearing stories, listener's brain activity mirrors the storyteller's—creating empathy and connection.
+
+**Dopamine Release:** Emotional stories trigger dopamine, improving memory and attention.
+
+**Oxytocin Production:** Character-driven stories release oxytocin, building trust.
+
+**Result:** Stories are 22x more memorable than facts alone.
+
+## Core Frameworks
+
+### 1. AIDA (Attention-Interest-Desire-Action)
+
+**Structure:**
+```
+A - Attention: Hook that stops the scroll
+I - Interest: Why this matters to them
+D - Desire: Paint the transformation
+A - Action: Clear next step
+```
+
+**Best For:** Short-form content, CTAs, promotional posts
+
+**Example Flow:**
+- A: "92% of creators make this mistake"
+- I: "It's costing them thousands of followers"
+- D: "Imagine doubling your reach in 30 days"
+- A: "Save this post to avoid the trap"
+
+### 2. PAS (Problem-Agitate-Solve)
+
+**Structure:**
+```
+P - Problem: Identify the pain point
+A - Agitate: Make it feel urgent
+S - Solve: Present your solution
+```
+
+**Best For:** Educational content, product positioning
+
+**Example Flow:**
+- P: "Struggling with low engagement?"
+- A: "Every ignored post is a missed opportunity. While you guess, competitors grow."
+- S: "Here's the exact framework that changed everything..."
+
+### 3. Hook-Story-Offer
+
+**Structure:**
+```
+Hook: Grab attention with emotion
+Story: Build connection through narrative
+Offer: Present the value/CTA
+```
+
+**Best For:** Personal brand content, sales
+
+**Example Flow:**
+- Hook: "I almost quit creating last month"
+- Story: "Here's what happened and what I discovered"
+- Offer: "Now I help others avoid the same trap"
+
+### 4. Hero's Journey (Condensed)
+
+**Structure:**
+```
+Ordinary World → Call to Adventure → Resistance
+↓
+Mentor/Discovery → Trials → Transformation
+↓
+Return with Wisdom
+```
+
+**Best For:** Transformation stories, long-form carousels
+
+### 5. Before-After-Bridge (BAB)
+
+**Structure:**
+```
+Before: Where they are now (pain)
+After: Where they could be (pleasure)
+Bridge: How to get there (your solution)
+```
+
+**Best For:** Transformation content, testimonials
+
+## Social Media Adaptations
+
+### For Instagram Carousels (7-10 slides)
+
+**Adapted Hero's Journey:**
+```
+S1: Ordinary World (hook with current state)
+S2: Call (the problem emerged)
+S3: Resistance (initial struggle)
+S4: Discovery (the turning point)
+S5-6: Trials (the process)
+S7: Transformation (the result)
+S8: Return (lessons for audience)
+S9: CTA
+```
+
+### For X/Twitter Threads (10 tweets)
+
+**Adapted PAS:**
+```
+T1: Hook + Problem statement
+T2-3: Agitate (expand on pain)
+T4: Transition ("But here's what I learned...")
+T5-8: Solution (one insight per tweet)
+T9: Summary
+T10: CTA + Retweet request
+```
+
+### For Reels/TikTok (60 seconds)
+
+**Compressed AIDA:**
+```
+0-3s: Attention (visual hook)
+3-15s: Interest (why watch)
+15-45s: Desire (value delivery)
+45-60s: Action (CTA)
+```
+
+## Framework Selection Guide
+
+| Content Goal | Best Framework | Why |
+|--------------|----------------|-----|
+| Teach something | PAS | Addresses pain before solution |
+| Share transformation | Hero's Journey | Full emotional arc |
+| Sell/promote | AIDA | Clear action path |
+| Build connection | Hook-Story-Offer | Personal narrative |
+| Show results | BAB | Clear contrast |
+
+## Implementation Tips
+
+### Start with the End
+Know your CTA before writing. Work backwards.
+
+### One Framework Per Piece
+Don't mix. Each framework has internal logic.
+
+### Emotion First, Logic Second
+Hook emotionally, support logically.
+
+### Test Different Frameworks
+Same topic, different structures. Track what resonates.

package/skills/skills/diverse-content-gen/SKILL.md

@@ -0,0 +1,201 @@
+---
+name: diverse-content-gen
+description: "Agent workflow for generating highly diverse creative content using Verbalized Sampling (VS) technique. Use when user requests multiple variations, brainstorming, creative ideas, or when standard prompting produces repetitive outputs. Increases diversity by 1.6-2.1× while maintaining quality. Works for: blog posts, social media captions, stories, campaign ideas, product descriptions, taglines, and open-ended creative tasks."
+category: creative
+---
+
+# Diverse Content Generation using Verbalized Sampling
+
+## Overview
+
+This skill teaches agents how to use **Verbalized Sampling (VS)** - a research-backed prompting technique that dramatically increases output diversity (1.6-2.1× improvement) without sacrificing quality.
+
+**The Problem:** Standard aligned LLMs suffer from "mode collapse" - they generate overly similar, safe, predictable outputs because of typicality bias in training data.
+
+**The Solution:** Instead of asking for single instances ("write a blog post"), VS prompts the model to verbalize a probability distribution over multiple responses ("generate 5 blog post ideas with their probabilities").
+
+**Core Principle:** Different prompt types collapse to different modes. Distribution-level prompts recover the diverse base model distribution, while instance-level prompts collapse to the most typical output.
+
+---
+
+## Workflow Decision Tree
+
+**Detect user intent, route to appropriate reference:**
+
+| User Request Pattern | Route To | Description |
+|---------------------|----------|-------------|
+| "Generate diverse [content]" | `references/vs-core-technique.md` | Learn VS basics, prompt templates, execution |
+| "Write 5 blog posts / captions / ideas" | `references/task-workflows.md` | Task-specific workflows pre-configured |
+| "Need higher quality" or "too wild" | `references/advanced-techniques.md` | VS-CoT, VS-Multi, parameter tuning |
+| "Save to file" or "batch process 50 items" | `references/tool-integration.md` | VS + File tools, batch workflows |
+| "VS outputs too similar" or errors | `references/troubleshooting.md` | Common pitfalls and solutions |
+| "Which model works best?" | `references/research-findings.md` | Benchmarks, model compatibility |
+
+**Default workflow:** Load `vs-core-technique.md` first, then load additional references as needed.
+
+---
+
+## When to Use This Skill
+
+### Trigger Scenarios
+
+**Use VS when user requests:**
+- "Give me **multiple variations/options/ideas**"
+- "I need **diverse** [content type]"
+- "**Brainstorm** several approaches to..."
+- "Generate **different angles** for..."
+- "Avoid repetitive/similar outputs"
+
+**Use VS for these content types:**
+- Creative writing (blog posts, stories, poems, scripts)
+- Marketing (campaign ideas, taglines, ad copy, social captions)
+- Product content (descriptions, feature bullets, value props)
+- Ideation (brainstorming, exploration, strategy options)
+- Open-ended QA (tasks with multiple valid answers)
+
+**DON'T use VS for:**
+- Single-answer factual questions
+- Tasks requiring deterministic output
+- When user explicitly wants "the best" single answer
+- Real-time low-latency applications
+
+---
+
+## Quick Start (30-Second Version)
+
+**For agents who need VS immediately:**
+
+### 1. Detect Need
+User wants multiple variations → Use VS
+
+### 2. Basic VS Prompt Template
+```
+Generate {k} responses to: {user_request}
+
+Return JSON format with key "responses" (list of dicts).
+Each dict must include:
+• text: the response string only
+• probability: estimated probability (0.0-1.0)
+
+Give ONLY the JSON object, no extra text.
+```
+
+### 3. Standard Parameters
+- **k = 5** (candidates per call)
+- **temperature = 0.8**
+- **threshold = 0.10** (optional, for more diversity)
+
+### 4. Process Output
+```python
+import json
+data = json.loads(llm_output)
+candidates = data["responses"]
+# Present to user ranked by probability
+```
+
+**For detailed instructions:** Load `references/vs-core-technique.md`
+
+---
+
+## Progressive Learning Path
+
+**Recommended loading sequence:**
+
+### Level 1: Basics (Required)
+1. Start here: `references/vs-core-technique.md`
+   - VS theory and why it works
+   - Copy-paste ready prompt templates
+   - Step-by-step execution workflow
+   - Output parsing and presentation
+
+### Level 2: Task-Specific (Choose based on use case)
+2. Load: `references/task-workflows.md`
+   - Blog post ideas workflow
+   - Social media captions workflow
+   - Campaign/strategy ideas workflow
+   - Story/narrative generation workflow
+
+### Level 3: Advanced (On-demand)
+3. When needed:
+   - **Higher quality needed:** `references/advanced-techniques.md` (VS-CoT, VS-Multi)
+   - **File operations:** `references/tool-integration.md` (Write, batch processing)
+   - **Issues/errors:** `references/troubleshooting.md` (Pitfalls & fixes)
+   - **Model selection:** `references/research-findings.md` (Benchmarks)
+
+---
+
+## Quick Reference Card
+
+**Copy this for quick lookup:**
+
+| Parameter | Default Value | When to Adjust |
+|-----------|--------------|----------------|
+| k (candidates) | 5 | Use 3 for quick, 10 for exploration |
+| Temperature | 0.7-1.0 | Combine with VS for extra diversity |
+| Probability threshold | 0.10 (optional) | Lower (0.01) for more creative outputs |
+
+**Troubleshooting shortcuts:**
+- Outputs too similar? → Lower threshold OR increase k OR load `advanced-techniques.md`
+- Quality too low? → VS-Multi workflow (see `advanced-techniques.md`)
+- JSON parsing errors? → Emphasize "ONLY JSON" OR use regex extraction
+- Not sure which model? → Load `research-findings.md`
+
+**Quality checklist before presenting:**
+- [ ] Diversity achieved (different angles/styles)
+- [ ] Quality maintained (baseline standards)
+- [ ] User intent matched
+- [ ] Clean formatting (no JSON artifacts)
+
+---
+
+## Resources
+
+This skill uses progressive disclosure for optimal token efficiency:
+
+### references/
+Documentation loaded on-demand based on agent needs:
+
+- **vs-core-technique.md** - Core VS concepts, prompt templates, execution steps
+- **task-workflows.md** - Pre-configured workflows for common content types
+- **advanced-techniques.md** - VS-CoT, VS-Multi, parameter tuning, refinement
+- **tool-integration.md** - Combining VS with file tools, batch processing
+- **troubleshooting.md** - Common pitfalls and solutions
+- **research-findings.md** - Performance benchmarks, model compatibility data
+
+**Pattern:** Agent loads SKILL.md first (routing), then loads specific references as needed during execution.
+
+---
+
+## Examples in Context
+
+### Example 1: Simple Brainstorming
+**User:** "Give me 5 tagline ideas for a coffee shop"
+
+**Agent workflow:**
+1. Detect: "5 ideas" → VS needed
+2. Load: `vs-core-technique.md` (if not already loaded)
+3. Execute: VS prompt with k=5
+4. Parse & present: 5 diverse taglines
+
+### Example 2: Production Content
+**User:** "Write 10 blog post ideas about AI, I need them saved to a file"
+
+**Agent workflow:**
+1. Detect: "10 ideas" + "saved to file" → VS + file tools
+2. Load: `vs-core-technique.md` + `tool-integration.md`
+3. Execute: VS with k=5, make 2 calls
+4. Process: Format as markdown
+5. Write: Use Write tool to save file
+
+### Example 3: Quality Refinement
+**User:** "These are good but need more polish for production use"
+
+**Agent workflow:**
+1. Detect: Quality improvement needed
+2. Load: `advanced-techniques.md`
+3. Execute: VS-Multi workflow (initial VS → user selects → refine)
+4. Deliver: Polished output
+
+---
+
+**Ready to start?** Load `references/vs-core-technique.md` to begin using VS.