@nockdev/awf 6.2.0 → 6.2.1

package/.agent/skills/ai-ml/gemini-tts/data/markup-tags.yaml

@@ -0,0 +1,95 @@
+# =============================================================================
+# markup-tags.yaml — TTS Markup Tags Reference
+# =============================================================================
+# 15 patterns for speech control via inline markup tags
+# =============================================================================
+
+version: "1.0.0"
+total_patterns: 15
+
+modes:
+  non_speech_sounds:
+    description: "Add natural non-verbal sounds"
+    tags:
+      - tag: "[laughing]"
+        desc: "Laughter"
+        example: "[laughing] That's the funniest thing I've heard all day!"
+      - tag: "[sigh]"
+        desc: "Exasperated or tired sigh"
+        example: "[sigh] I suppose we'll have to start over."
+      - tag: "[gasp]"
+        desc: "Surprise or shock"
+        example: "[gasp] I had no idea!"
+      - tag: "[clears throat]"
+        desc: "Throat clearing"
+        example: "[clears throat] Allow me to begin."
+      - tag: "[sniffles]"
+        desc: "Sniffling"
+        example: "[sniffles] It's just so beautiful."
+
+  speaking_style:
+    description: "Modify vocal delivery style"
+    tags:
+      - tag: "[whispering]"
+        desc: "Quiet, secretive tone"
+        example: "[whispering] I have something important to tell you..."
+      - tag: "[sarcasm]"
+        desc: "Sarcastic delivery"
+        example: "[sarcasm] Oh wow, what a brilliant plan."
+      - tag: "[cheerfully]"
+        desc: "Happy, upbeat tone"
+        example: "[cheerfully] Great news everyone!"
+      - tag: "[sadly]"
+        desc: "Somber, melancholic tone"
+        example: "[sadly] I'm afraid it's too late."
+      - tag: "[excitedly]"
+        desc: "Energetic, enthusiastic"
+        example: "[excitedly] You won't believe what happened!"
+      - tag: "[calmly]"
+        desc: "Peaceful, measured delivery"
+        example: "[calmly] Take a deep breath and relax."
+
+  vocalized_sounds:
+    description: "Musical and vocalized expressions"
+    tags:
+      - tag: "[singing]"
+        desc: "Singing voice"
+        example: "[singing] Happy birthday to you..."
+      - tag: "[humming]"
+        desc: "Humming melody"
+        example: "[humming] Hmm hmm hmm..."
+
+  pacing_pronunciation:
+    description: "Control speech speed and timing"
+    tags:
+      - tag: "[extremely fast]"
+        desc: "Rapid speech delivery"
+        example: "[extremely fast] And the winner of tonight's grand prize is..."
+      - tag: "[slowly]"
+        desc: "Deliberate, slow delivery"
+        example: "[slowly] Let me explain this one more time..."
+      - tag: "[pause: Xs]"
+        desc: "Insert silence (X seconds)"
+        example: "The answer is... [pause: 2s] ...forty-two."
+
+usage_tips:
+  - "Tags go BEFORE the text they modify"
+  - "Tags affect the sentence/phrase immediately following"
+  - "Combine tags: '[whispering][slowly] This is very important...'"
+  - "Tags are case-insensitive: [LAUGHING] = [laughing]"
+  - "Not all tags work equally well — experiment with model"
+  - "Pro model handles markup most naturally"
+
+combinations:
+  example_dialogue: |
+    [cheerfully] Welcome to today's episode!
+    [pause: 1s]
+    [calmly] Today we're going to discuss something fascinating.
+    [excitedly] You won't believe the results we found!
+    [whispering] But first, let me share a little secret...
+    [laughing] Just kidding, it's not that dramatic.
+    [sigh] Anyway, let's get started.
+
+# =============================================================================
+# _DOMYH Awesome Code • Markup Tags v1.0_
+# =============================================================================

package/.agent/skills/ai-ml/gemini-tts/data/models.yaml

@@ -0,0 +1,124 @@
+# =============================================================================
+# models.yaml — Gemini TTS Model Comparison
+# =============================================================================
+# 6 patterns for model selection, voice options, output configuration
+# Source: https://docs.cloud.google.com/text-to-speech/docs/gemini-tts
+# =============================================================================
+
+version: "1.0.0"
+total_patterns: 6
+
+models:
+  flash:
+    id: "gemini-2.5-flash-preview-tts"
+    display: "Gemini TTS Flash"
+    tier: "speed"
+    latency: "lowest"
+    quality: "good"
+    cost: "low"
+    streaming: true
+    multi_speaker: true
+    best_for:
+      - "Real-time applications"
+      - "Interactive voice agents"
+      - "Live narration"
+
+  lite:
+    id: "gemini-2.5-flash-lite-preview-tts"
+    display: "Gemini TTS Lite"
+    tier: "cost"
+    latency: "low"
+    quality: "good"
+    cost: "lowest"
+    streaming: true
+    multi_speaker: true
+    best_for:
+      - "High-volume batch processing"
+      - "Cost-sensitive applications"
+      - "Notifications and alerts"
+
+  pro:
+    id: "gemini-2.5-pro-preview-tts"
+    display: "Gemini TTS Pro"
+    tier: "quality"
+    latency: "higher"
+    quality: "best"
+    cost: "higher"
+    streaming: true
+    multi_speaker: true
+    best_for:
+      - "Audiobooks"
+      - "Premium voiceovers"
+      - "Marketing content"
+      - "Podcast production"
+
+voices:
+  description: "Prebuilt voice options available across all models"
+  options:
+    - name: "Kore"
+      style: "Clear, professional female"
+    - name: "Puck"
+      style: "Warm, conversational male"
+    - name: "Charon"
+      style: "Deep, authoritative male"
+    - name: "Aoede"
+      style: "Bright, energetic female"
+    - name: "Fenrir"
+      style: "Calm, measured male"
+    - name: "Leda"
+      style: "Friendly, approachable female"
+
+output_formats:
+  - mime: "audio/wav"
+    desc: "WAV (PCM 24kHz)"
+    use_case: "High quality editing"
+  - mime: "audio/ogg;codecs=opus"
+    desc: "OGG Opus"
+    use_case: "Streaming, low bandwidth"
+  - mime: "audio/mp3"
+    desc: "MP3"
+    use_case: "Universal compatibility"
+  - mime: "audio/flac"
+    desc: "FLAC"
+    use_case: "Lossless archival"
+  - mime: "audio/aac"
+    desc: "AAC"
+    use_case: "Mobile playback"
+  - mime: "audio/L16;rate=24000"
+    desc: "Raw PCM 24kHz"
+    use_case: "Real-time processing"
+
+api_usage:
+  basic:
+    python: |
+      from google import genai
+      client = genai.Client(api_key="API_KEY")
+      response = client.models.generate_content(
+          model="gemini-2.5-flash-preview-tts",
+          contents="Hello, welcome to our product demo!",
+          config=genai.types.GenerateContentConfig(
+              response_modalities=["AUDIO"],
+              speech_config=genai.types.SpeechConfig(
+                  voice_config=genai.types.VoiceConfig(
+                      prebuilt_voice_config=genai.types.PrebuiltVoiceConfig(
+                          voice_name="Kore"
+                      )
+                  )
+              )
+          )
+      )
+      # Save: response.candidates[0].content.parts[0].inline_data.data
+
+  streaming:
+    python: |
+      for chunk in client.models.generate_content_stream(
+          model="gemini-2.5-flash-preview-tts",
+          contents="Long text to narrate...",
+          config=config
+      ):
+          # Process each audio chunk in real-time
+          audio_data = chunk.candidates[0].content.parts[0].inline_data.data
+
+# =============================================================================
+# _DOMYH Awesome Code • TTS Models v1.0_
+# =============================================================================

package/.agent/skills/ai-ml/gemini-tts/data/prompting-patterns.yaml

@@ -0,0 +1,81 @@
+# =============================================================================
+# prompting-patterns.yaml — TTS Style Prompting Patterns
+# =============================================================================
+# 8 patterns for controlling TTS output via natural language prompts
+# =============================================================================
+
+version: "1.0.0"
+total_patterns: 8
+
+style_prompts:
+  description: "Use natural language to direct voice style"
+  principle: "System instruction sets the voice persona and delivery style"
+
+  patterns:
+    news_anchor:
+      prompt: "Read this in a professional, authoritative news anchor voice with clear enunciation."
+      use_case: "News summaries, announcements"
+
+    podcast_host:
+      prompt: "Read this in a warm, conversational podcast host voice. Be engaging and natural."
+      use_case: "Podcast scripts, casual narration"
+
+    storyteller:
+      prompt: "Read this like a bedtime story narrator — gentle, expressive, with dramatic pauses."
+      use_case: "Audiobooks, children's content"
+
+    instructor:
+      prompt: "Read this like a patient, clear teacher explaining a concept. Emphasize key terms."
+      use_case: "Educational content, tutorials"
+
+    dramatic:
+      prompt: "Read this with dramatic flair, varying pace and intensity for emotional impact."
+      use_case: "Theater scripts, dramatic readings"
+
+    meditation:
+      prompt: "Read this in a calm, soothing meditation guide voice. Slow pace, soft tone."
+      use_case: "Guided meditation, relaxation content"
+
+    sports_commentator:
+      prompt: "Read this with the energy and excitement of a live sports commentator."
+      use_case: "Sports updates, exciting announcements"
+
+    documentary:
+      prompt: "Read this in a BBC documentary narrator style — informative, measured, with gravitas."
+      use_case: "Documentary narration, educational videos"
+
+emotionally_rich_text:
+  description: "Write text that naturally triggers expressive delivery"
+  tips:
+    - "Use punctuation for natural pauses: commas, ellipses, em dashes"
+    - "Exclamation marks increase energy: 'Amazing!' vs 'Amazing.'"
+    - "Question marks trigger interrogative intonation"
+    - "Longer sentences → more natural flow"
+    - "Short sentences → punchy, impactful delivery"
+
+  example_text: |
+    Have you ever wondered... what makes a great voice?
+    It's not just about clarity — it's about emotion.
+    The way a pause... can build anticipation.
+    The way excitement can make your heart race!
+    And sometimes, a whisper is the most powerful tool of all.
+
+alignment_strategies:
+  description: "Ensure voice style matches content tone"
+  rules:
+    - "Match prompt style to content genre (don't use news voice for poetry)"
+    - "Use markup tags for fine-grained moments within a consistent style"
+    - "Test with short segments before full narration"
+    - "Pro model best for style adherence"
+    - "Multi-speaker: assign different styles per speaker"
+
+multi_language:
+  description: "TTS supports multiple languages"
+  tips:
+    - "Specify language in prompt: 'Read in Vietnamese with natural pronunciation'"
+    - "Model auto-detects language from text"
+    - "Code-switching (mixed languages) works but quality varies"
+
+# =============================================================================
+# _DOMYH Awesome Code • TTS Prompting Patterns v1.0_
+# =============================================================================

package/.agent/skills/ai-ml/prompt-engineering/META.yaml

@@ -0,0 +1,77 @@
+name: prompt-engineering
+version: "2.0.0"
+display: Prompt Engineering Patterns
+category: ai-ml
+tier: 1
+priority: 1
+
+desc: "Prompt engineering for LLMs: chain-of-thought, few-shot, ReAct, tree-of-thought, structured output, system prompts, guardrails. Gemini 3 agentic 9-rule template, thinking controls, function calling, MCP. Multi-provider patterns (OpenAI, Anthropic, Google). 155+ patterns."
+
+triggers:
+  file_patterns: []
+  keywords:
+    - prompt
+    - prompt engineering
+    - chain of thought
+    - few-shot
+    - zero-shot
+    - system prompt
+    - llm
+    - ai
+    - instruction
+    - reasoning
+    - react pattern
+    - tree of thought
+    - structured output
+    - guardrail
+    - jailbreak
+    # Gemini 3 / v2.0
+    - gemini 3
+    - agentic template
+    - thinking
+    - thinking budget
+    - function calling
+    - compositional function
+    - MCP function calling
+    - grounding
+    - context caching
+  intents:
+    - write prompts
+    - prompt design
+    - LLM integration
+    - AI instruction
+    - agentic system design
+
+caps:
+  - Chain-of-thought (CoT) prompting
+  - Few-shot / zero-shot learning
+  - ReAct (Reasoning + Acting)
+  - Tree-of-thought (ToT)
+  - Structured output (JSON mode, Pydantic, Zod)
+  - System prompt design
+  - Prompt injection defense
+  - Multi-turn conversation design
+  - Evaluation and testing prompts
+  - Tool use / function calling
+  - RAG prompt patterns
+  - Multi-provider patterns (OpenAI, Anthropic, Google)
+  # v2.0
+  - Gemini 3 agentic 9-rule template
+  - Thinking controls (levels, budgets, summaries)
+  - Function calling (parallel, compositional, MCP)
+  - Context caching strategies
+  - Grounding with Google Search
+
+related_skills:
+  - ai-agents
+  - gemini-live
+  - gemini-media-gen
+
+data_files:
+  - data/reasoning-patterns.yaml
+  - data/output-patterns.yaml
+  - data/safety-patterns.yaml
+  - data/provider-patterns.yaml
+  # v2.0
+  - data/gemini3-patterns.yaml
+  - data/tool-patterns.yaml

package/.agent/skills/ai-ml/prompt-engineering/SKILL.md

@@ -0,0 +1,217 @@
+# Prompt Engineering Patterns v2.0
+
+> 🧠 **LLM prompt design for production AI systems**
+> **Patterns**: 155+ | **Providers**: 3 | **Categories**: 6
+
+---
+
+## Quick Reference
+
+| What You Need                           | Data File                 | Patterns |
+| --------------------------------------- | ------------------------- | -------- |
+| Reasoning (CoT, ReAct, ToT)             | `reasoning-patterns.yaml` | 35       |
+| Output (JSON, structured, formatting)   | `output-patterns.yaml`    | 30       |
+| Safety (injection, guardrails)          | `safety-patterns.yaml`    | 30       |
+| Provider-specific (OpenAI, Claude, etc) | `provider-patterns.yaml`  | 25       |
+| **Gemini 3 (agentic, thinking, XML)**   | `gemini3-patterns.yaml`   | 20       |
+| **Tool use (FC, MCP, multi-tool)**      | `tool-patterns.yaml`      | 15       |
+
+---
+
+## Prompting Hierarchy
+
+```
+┌──────────────────────────────────────────┐
+│  Level 1: Zero-Shot (no examples)        │
+│  Level 2: Few-Shot (1-5 examples)        │
+│  Level 3: Chain-of-Thought (reasoning)   │
+│  Level 4: ReAct (reasoning + tools)      │
+│  Level 5: Tree-of-Thought (exploration)  │
+│  Level 6: Multi-agent (delegation)       │
+│  Level 7: Agentic (Gemini 3 template)    │ ← NEW
+└──────────────────────────────────────────┘
+```
+
+---
+
+## Core Techniques
+
+| Technique            | When to Use                  | Accuracy |
+| -------------------- | ---------------------------- | -------- |
+| **Zero-shot**        | Simple tasks, strong model   | ~70%     |
+| **Few-shot**         | Pattern matching, formatting | ~80%     |
+| **Chain-of-thought** | Math, logic, multi-step      | ~90%     |
+| **Self-consistency** | High-stakes decisions        | ~92%     |
+| **ReAct**            | Tool use, research           | ~88%     |
+| **Tree-of-thought**  | Creative/exploration tasks   | ~85%     |
+
+---
+
+## Gemini 3 Agentic Template (NEW)
+
+Google's official 9-rule agentic system instruction template:
+
+```xml
+<role>You are [role]. Your goal is [goal].</role>
+
+<instructions>
+1. [Primary instruction]
+2. [Secondary instruction]
+...
+</instructions>
+
+<constraints>
+- [Constraint 1]
+- [Constraint 2]
+</constraints>
+
+<output_format>
+[Expected format: JSON, markdown, etc.]
+</output_format>
+
+<context>
+[Background information, domain knowledge]
+</context>
+
+<task>
+[Specific task to complete]
+</task>
+```
+
+**Best Practices**:
+
+- Use XML tags for structured sections
+- Positive patterns > negative patterns ("do X" not "don't do Y")
+- Put examples WITHIN output_format for few-shot
+- Completion strategy gives output prefix to guide direction
+
+---
+
+## Thinking Controls (NEW)
+
+| Level    | Name        | Use Case                    | Budget          |
+| -------- | ----------- | --------------------------- | --------------- |
+| `off`    | No thinking | Simple lookups              | 0               |
+| `low`    | Quick       | Classification, formatting  | 1K-2K           |
+| `medium` | Balanced    | General tasks               | 4K-8K           |
+| `high`   | Deep        | Math, code analysis         | 16K-32K         |
+| `max`    | Maximum     | Research, complex reasoning | Up to model max |
+
+```python
+# Thinking level (Gemini 3)
+response = model.generate_content(
+    "prompt",
+    generation_config={"thinking_config": {"thinking_level": "high"}}
+)
+
+# Thinking budget (token limit)
+response = model.generate_content(
+    "prompt",
+    generation_config={"thinking_config": {"thinking_budget": 8192}}
+)
+```
+
+---
+
+## Structured Output (NEW)
+
+| Pattern                | Schema Tool                            | When to Use           |
+| ---------------------- | -------------------------------------- | --------------------- |
+| **Pydantic** (Python)  | `response_schema=Recipe`               | Type-safe Python apps |
+| **Zod** (JavaScript)   | `responseMimeType: "application/json"` | Node.js/TS apps       |
+| **JSON Schema** (REST) | `responseSchema: {...}`                | Direct API calls      |
+| **Enum**               | `enum: ["cat", "dog"]`                 | Classification tasks  |
+
+**Best Practices**:
+
+- Use `description` field in schema for model guidance
+- Strong typing (`integer`, `enum`) > loose typing (`string`)
+- Validate semantically even if syntactically correct
+
+---
+
+## Function Calling Patterns (NEW)
+
+| Pattern                | Description                       | Use Case                              |
+| ---------------------- | --------------------------------- | ------------------------------------- |
+| **Single**             | One function declaration          | Simple API calls                      |
+| **Parallel**           | Multiple functions at once        | Independent data fetching             |
+| **Compositional**      | Chained function results          | `get_location()` → `get_weather(loc)` |
+| **MCP Built-in**       | Gemini SDK auto-calls MCP tools   | MCP server integration                |
+| **Multi-tool**         | Combine Search + Code + Functions | Complex agentic workflows             |
+| **Automatic** (Python) | SDK handles full loop             | Rapid prototyping                     |
+
+**Modes**: `AUTO` (default) | `ANY` (force call) | `NONE` (disable)
+
+---
+
+## Prompt Structure
+
+```
+[System Prompt]
+  - Role definition
+  - Capabilities and limitations
+  - Output format requirements
+  - Safety guardrails
+
+[User Prompt]
+  - Context/background
+  - Specific task description
+  - Input data
+  - Expected output format
+  - Examples (few-shot)
+
+[Assistant prefill] (Claude-specific)
+  - Guide initial response direction
+```
+
+---
+
+## Chain-of-Thought Patterns
+
+| Pattern              | Description                                  |
+| -------------------- | -------------------------------------------- |
+| **Basic CoT**        | "Think step by step"                         |
+| **Zero-shot CoT**    | "Let's think about this carefully"           |
+| **Few-shot CoT**     | Provide examples WITH reasoning steps        |
+| **Self-consistency** | Generate N answers, pick majority            |
+| **Auto-CoT**         | Let model generate its own reasoning prompts |
+
+---
+
+## Output Formatting
+
+| Format       | Prompt Pattern                                             |
+| ------------ | ---------------------------------------------------------- |
+| **JSON**     | "Respond in valid JSON matching this schema: {...}"        |
+| **Markdown** | "Format as markdown with headers and bullet points"        |
+| **Code**     | "Write [language] code with comments explaining each step" |
+| **Table**    | "Present results in a markdown table with columns: [...]"  |
+| **XML**      | "Wrap your response in <answer>...</answer> tags"          |
+
+---
+
+## Safety Patterns
+
+| Threat               | Defense                                        |
+| -------------------- | ---------------------------------------------- |
+| **Prompt injection** | Input sanitization + delimiter tags            |
+| **Jailbreaking**     | Constitutional AI rules + refusal training     |
+| **Data leakage**     | PII filtering + output validation              |
+| **Hallucination**    | RAG grounding + citation requirements          |
+| **Bias**             | Balanced examples + explicit fairness criteria |
+
+---
+
+## HSA Integration
+
+Data powered by HSA BM25 search engine. Query YAML data via skill search:
+
+| Domain    | Query Examples                            |
+| --------- | ----------------------------------------- |
+| Reasoning | "chain of thought step by step reasoning" |
+| Output    | "JSON schema structured output format"    |
+| Safety    | "prompt injection defense guardrail"      |
+| Provider  | "OpenAI function calling Claude tool use" |
+| Gemini 3  | "agentic template thinking budget XML"    |
+| Tools     | "MCP compositional parallel function"     |