@shaykec/bridge 0.4.25 → 0.4.26

package/modules/prioritization-frameworks/quick-ref.md

@@ -0,0 +1,77 @@
+# Prioritization Quick Reference
+
+## RICE Formula
+
+```
+RICE = (Reach × Impact × Confidence) ÷ Effort
+```
+
+| Factor | Type | Scale / Units |
+|--------|------|---------------|
+| Reach | Number | Users, events, or conversions per timeframe |
+| Impact | 0.25–3 | 0.25 min, 0.5 low, 1 med, 2 high, 3 massive |
+| Confidence | % | 50%, 80%, 100% |
+| Effort | Person-months | Total team capacity |
+
+**Higher RICE = higher priority.**
+
+## MoSCoW Buckets
+
+| Bucket | Meaning | Guideline |
+|--------|----------|-----------|
+| **M**ust | Launch blockers | ≤ 60% of scope |
+| **S**hould | Important, not blocking | |
+| **C**ould | Nice to have | If time permits |
+| **W**on't | Out of scope | Capture for later |
+
+## Impact Mapping Structure
+
+```
+Goal: [Business outcome]
+├── Actor: [Who can help/block]
+│   ├── Impact: [Behavior change needed]
+│   │   ├── Deliverable: [What we build]
+│   │   └── Deliverable: [...]
+│   └── Impact: [...]
+└── Actor: [...]
+```
+
+## Value vs Effort Matrix
+
+| | Low Effort | High Effort |
+|---|------------|-------------|
+| **High Value** | Quick wins ✓ | Big bets |
+| **Low Value** | Fill-ins | Avoid ✗ |
+
+## Framework Comparison
+
+| Use Case | Best Framework |
+|----------|----------------|
+| Ranked backlog, quantitative | RICE |
+| Release scope, stakeholder alignment | MoSCoW |
+| Goal alignment, assumption surfacing | Impact Mapping |
+| Quick triage | Value/Effort |
+
+## Communicating Priorities
+
+1. **Show the model** — Don't hide; share how you scored
+2. **Make inputs editable** — Disagreement? Change inputs, re-run
+3. **Capture "Won't"** — Explicit out-of-scope prevents surprises
+4. **Revisit often** — New info → new scores
+
+## RICE Example
+
+| Initiative | R | I | C | E | RICE |
+|------------|---|---|---|---|------|
+| Dark mode | 10k | 0.5 | 100% | 1 | 5,000 |
+| Search | 8k | 2 | 80% | 3 | 4,267 |
+| Export | 2k | 1 | 100% | 0.5 | 4,000 |
+
+## Impact Map (Mermaid)
+
+```mermaid
+flowchart TB
+  G[Goal] --> A[Actor]
+  A --> I[Impact]
+  I --> D[Deliverable]
+```

package/modules/prioritization-frameworks/quiz.md

@@ -0,0 +1,73 @@
+# Prioritization — Quiz
+
+## Question 1
+
+In the RICE formula, which factor is the denominator?
+
+A) Reach
+B) Impact
+C) Confidence
+D) Effort
+
+<!-- ANSWER: D -->
+<!-- EXPLANATION: RICE = (Reach × Impact × Confidence) ÷ Effort. Reach, Impact, and Confidence are the "benefit" factors (multiplied); Effort is the "cost" (divisor). Higher effort lowers the score. -->
+
+## Question 2
+
+In MoSCoW prioritization, what is the main risk of having too many "Must have" items?
+
+A) The document becomes too long
+B) Nothing is truly prioritized—everything is "critical"
+C) Stakeholders will reject the plan
+D) Engineering estimates become inaccurate
+
+<!-- ANSWER: B -->
+<!-- EXPLANATION: If everything is Must, nothing is. Must should be reserved for true launch-blockers. A common guideline is to cap Must at ~60% of scope. Too many Musts means you're either over-scoped or under-prioritized. -->
+
+## Question 3
+
+In Impact Mapping, what does "Impact" represent?
+
+A) The number of users affected
+B) A behavior change required of an actor to support the goal
+C) The revenue generated by a deliverable
+D) The effort to build a feature
+
+<!-- ANSWER: B -->
+<!-- EXPLANATION: Impact is the middle layer: Goal → Actor → Impact → Deliverable. Impact answers: "How must this actor change their behavior to help us achieve the goal?" It's about cause and effect, not scope or effort. -->
+
+## Question 4
+
+A feature has Reach 4,000, Impact 2, Confidence 80%, and Effort 2. What is its RICE score?
+
+A) 3,200
+B) 1,280
+C) 2,560
+D) 6,400
+
+<!-- ANSWER: A -->
+<!-- EXPLANATION: RICE = (Reach × Impact × Confidence) ÷ Effort = (4000 × 2 × 0.8) ÷ 2 = 6400 ÷ 2 = 3200. -->
+
+## Question 5
+
+Which framework is best for "what's in this release vs next release"?
+
+A) RICE
+B) MoSCoW
+C) Impact Mapping
+D) Value vs Effort matrix
+
+<!-- ANSWER: B -->
+<!-- EXPLANATION: MoSCoW explicitly buckets work into Must/Should/Could/Won't, which maps cleanly to release scope. RICE ranks but doesn't define release boundaries. Impact Mapping is strategic; Value/Effort is a triage view. MoSCoW is designed for scope negotiation. -->
+
+## Question 6
+
+In a Value vs Effort matrix, where should "quick wins" go?
+
+A) Low value, low effort
+B) High value, high effort
+C) High value, low effort
+D) Low value, high effort
+
+<!-- ANSWER: C -->
+<!-- EXPLANATION: Quick wins are high value with low effort—they deliver impact without huge cost. Do these first. Low value + high effort = avoid. High value + high effort = big bets. Low value + low effort = fill-ins. -->

package/modules/prioritization-frameworks/resources.md

@@ -0,0 +1,32 @@
+# Prioritization — Resources
+
+## Videos
+
+- [RICE Prioritization Framework Explained](https://www.youtube.com/watch?v=8Q6o-3LbQPE) — Product Manager HQ, ~12 min. Step-by-step RICE walkthrough.
+- [Impact Mapping in 15 Minutes](https://www.youtube.com/watch?v=8Q6o-3LbQPE) — Gojko Adzic. Quick intro to the impact mapping technique.
+
+## Articles and Readings
+
+- [RICE Scoring Model](https://www.productplan.com/glossary/rice-scoring-model/) — ProductPlan. Official RICE overview: formula, factors, and examples.
+- [Where do product roadmaps come from?](https://medium.com/intercom-inside/where-do-product-roadmaps-come-from-b526a9c60493) — Intercom. Origin of RICE and roadmap thinking.
+- [How to Use MoSCoW Prioritization](https://www.productplan.com/glossary/moscow-prioritization/) — ProductPlan. MoSCoW explained with examples.
+- [Impact Mapping](https://www.impactmapping.org/) — Gojko Adzic. The official impact mapping site with guides and examples.
+- [How to get the most out of impact mapping](https://gojko.net/2014/11/17/how-to-get-the-most-out-of-impact-mapping) — Gojko Adzic. Practical tips and anti-patterns.
+- [Prioritization Frameworks for Product Managers](https://www.productplan.com/product-management-frameworks/) — ProductPlan. Comparison of RICE, MoSCoW, value scoring, and more.
+
+## Books
+
+- **Impact Mapping** by Gojko Adzic — The canonical guide to impact mapping; short, practical.
+- **Inspired** by Marty Cagan — SVPG perspective on opportunity assessment and prioritization.
+- **Escaping the Build Trap** by Melissa Perri — Prioritization in the context of outcome-driven product management.
+
+## Podcasts
+
+- [Lenny's Podcast: How to prioritize](https://www.lennyspodcast.com/) — Lenny Rachitsky. Episodes on roadmap and prioritization with PM leaders.
+- [Product Talk](https://www.producttalk.org/) — Teresa Torres. Opportunity solution trees and evidence-based prioritization.
+
+## Tools
+
+- [ProductPlan](https://www.productplan.com/) — Roadmapping and prioritization software with RICE built-in.
+- [Aha!](https://www.aha.io/) — Product strategy and roadmap tools with scoring frameworks.
+- [Miro Impact Mapping Template](https://miro.com/templates/impact-mapping/) — Collaborative impact mapping canvas.

package/modules/prioritization-frameworks/walkthrough.md

@@ -0,0 +1,69 @@
+# Prioritization Walkthrough — Learn by Doing
+
+## Before We Begin
+
+**Diagnostic:** Why prioritize at all? If you could only ship *one* initiative this quarter, how would you choose which one? What would "the right choice" mean in your context?
+
+**Checkpoint:** You can name at least one trade-off you'd face (e.g., revenue vs. retention, quick win vs. strategic bet) and why a gut feel isn't enough.
+
+---
+
+## Step 1: Compute RICE Scores
+
+<!-- hint:diagram mermaid-type="flowchart" topic="RICE formula: Reach × Impact × Confidence ÷ Effort" -->
+
+**Task:** Three initiatives: (A) Reach 5,000, Impact 2, Confidence 80%, Effort 2 person-months. (B) Reach 3,000, Impact 1, Confidence 100%, Effort 0.5. (C) Reach 8,000, Impact 0.5, Confidence 50%, Effort 4. Compute RICE for each. Rank them.
+
+**Question:** Why does (B) rank highest despite lower Reach? What would change if you doubled (C)'s Confidence?
+
+**Checkpoint:** The user computes: A ≈ 4,000, B ≈ 6,000, C ≈ 500. Rank: B, A, C. They understand Effort and Confidence strongly affect the score.
+
+---
+
+## Step 2: Apply MoSCoW to a Release
+
+**Task:** You're planning a "v2 search" release. Assign these to M/S/C/W (max 3 Must, max 3 Should): basic keyword search, filters, search suggestions, export results, save searches, analytics dashboard, API for search, keyboard shortcuts.
+
+**Question:** What's the risk of making "export results" a Must? How would you negotiate if stakeholders want 5 Musts?
+
+**Checkpoint:** The user assigns items with justification. They recognize basic search is Must; export might be Should or Could. They can explain why capping Must forces tradeoffs.
+
+---
+
+## Step 3: Build an Impact Map
+
+**Task:** Goal: "Reduce support tickets about 'how do I reset my password' by 30% in Q3." Build an impact map: 2 actors, 2 impacts per actor, 2 deliverables per impact.
+
+**Question:** Why does Impact describe *behavior change* rather than *features*? What assumption are you making?
+
+**Checkpoint:** The user produces: actors (e.g., users, support agents), impacts (e.g., "users self-serve reset"), deliverables (e.g., clearer UX, reset email). They state at least one assumption (e.g., "users will use the new flow").
+
+---
+
+## Step 4: Plot Value vs Effort
+
+**Task:** Take 6 real or hypothetical initiatives from your work. Define "value" and "effort" for your context. Plot each on a 2×2. Label: 1 quick win, 1 big bet, 1 avoid.
+
+**Question:** Who might disagree with your "avoid" placement? How would you defend it?
+
+**Checkpoint:** The user has a matrix with 6 items. Quick win = high value, low effort. Avoid = low value, high effort. They can justify the avoid and anticipate stakeholder pushback.
+
+---
+
+## Step 5: Defend a Priority
+
+**Task:** You prioritized "Dark mode" over "Export to PDF" using RICE. A stakeholder says: "Export is critical for enterprise users." Write your response in 4–6 sentences. Reference the model; acknowledge their concern; offer a path forward.
+
+**Question:** What would you do if they provide new data (e.g., enterprise is 40% of revenue) that changes Reach or Impact?
+
+**Checkpoint:** The user's response shows the numbers, explains why Dark mode scored higher, acknowledges the concern, and offers to re-score with updated inputs or add Export to next quarter.
+
+---
+
+## Step 6: Combine Frameworks
+
+**Task:** For a single product initiative, use both RICE and Impact Mapping. First: write the goal and build a 2-layer impact map. Then: pick 3 deliverables from the map and score them with RICE. What does each framework reveal that the other doesn't?
+
+**Question:** When would you lead with Impact Mapping vs RICE in a stakeholder meeting? Why?
+
+**Checkpoint:** The user produces an impact map and RICE scores for 3 deliverables. They articulate: Impact Mapping shows *why* and *who*; RICE provides *rank order*. They can choose the right lead depending on audience (strategic vs tactical).

package/modules/prompt-engineering/content.md

@@ -0,0 +1,123 @@
+# Prompt Engineering — Techniques That Get Better AI Outputs
+
+<!-- hint:slides topic="Prompt engineering: prompt anatomy, zero-shot vs few-shot, chain-of-thought, structured output, and system prompts" slides="5" -->
+
+## What Makes a Good Prompt?
+
+Prompts are instructions you give to an LLM. A good prompt is **specific**, **structured**, and **contextual** — it tells the model exactly what you want, in what format, with enough context to succeed.
+
+## Anatomy of a Prompt
+
+Every effective prompt can be broken into five components:
+
+```mermaid
+flowchart TB
+    subgraph prompt["Prompt Anatomy"]
+        R[Role]
+        C[Context]
+        T[Task]
+        F[Format]
+        X[Constraints]
+    end
+    R --> C --> T --> F --> X
+    X --> Out[Better Output]
+```
+
+| Component | Purpose | Example |
+|-----------|---------|---------|
+| **Role** | Sets the AI's persona and expertise | "You are an expert Python developer." |
+| **Context** | Background the model needs | "The codebase uses FastAPI and Pydantic." |
+| **Task** | The specific request | "Add validation for the email field." |
+| **Format** | Desired output structure | "Return a JSON object with keys: valid, errors." |
+| **Constraints** | Limits and rules | "Keep under 100 words. No external APIs." |
+
+## Zero-Shot vs Few-Shot Prompting
+
+**Zero-shot** — No examples. The model infers from the instruction alone.
+
+```
+Translate "Hello" to Spanish.
+```
+
+**Few-shot** — Provide 1–5 examples to establish the pattern.
+
+```
+Translate to Spanish:
+Hello → Hola
+Goodbye → Adiós
+Please → Por favor
+Thank you → ?
+```
+
+Few-shot dramatically improves accuracy for structured, style-sensitive, or edge-case tasks. The model learns the pattern from examples.
+
+## Chain-of-Thought Reasoning
+
+Ask the model to "think step by step" or "show your reasoning." This improves performance on logic, math, and multi-step tasks.
+
+```
+What is 17 × 24?
+
+Think through each step before giving the final answer.
+```
+
+CoT elicits intermediate reasoning, reducing errors and making output verifiable.
+
+## System Prompts
+
+Many APIs support a **system prompt** — instructions that apply across the conversation, separate from user messages. Use it for:
+
+- Role and persona
+- Output format preferences
+- Guardrails and constraints
+- Domain knowledge
+
+## Structured Output
+
+Request specific formats to make output machine-readable:
+
+- **JSON** — `Return a JSON object with: name, email, score`
+- **XML** — `Wrap the summary in <summary> tags`
+- **Markdown** — `Format as a markdown table with columns A, B, C`
+
+## Iterating on Prompts
+
+1. **Start simple** — one clear task
+2. **Add context** — if output is generic or wrong
+3. **Add examples** — if the format or style drifts
+4. **Add constraints** — if output is too long or off-topic
+5. **Test edge cases** — ambiguous inputs, empty inputs
+
+## Temperature and Creativity
+
+- **Low (0–0.3)** — Deterministic, factual, consistent. Good for code, extraction, classification.
+- **High (0.7–1.0)** — Creative, varied. Good for brainstorming, stories, varied phrasing.
+
+## Common Mistakes
+
+| Mistake | Fix |
+|---------|-----|
+| Vague instructions | Be explicit: "List 3 pros and 3 cons" |
+| No examples | Add 1–2 few-shot examples for format/style |
+| Too many constraints at once | Add constraints incrementally |
+| Assuming the model knows your context | Provide relevant background |
+| One giant prompt | Break into steps or sub-tasks |
+
+## Prompting by Task Type
+
+| Task | Strategy |
+|------|----------|
+| **Summarization** | Specify length, key points, audience |
+| **Analysis** | Define criteria, structure (pros/cons, compare/contrast) |
+| **Code** | Language, framework, patterns, input/output format |
+| **Creative writing** | Tone, length, style, examples of desired voice |
+
+---
+
+## Key Takeaways
+
+1. **Structure your prompts** — role, context, task, format, constraints
+2. **Few-shot beats zero-shot** for structured or style-sensitive tasks
+3. **Chain-of-thought** helps with reasoning and multi-step problems
+4. **Iterate** — start simple, add context and examples as needed
+5. **Avoid** vague instructions, no examples, and constraint overload

package/modules/prompt-engineering/exercises.md

@@ -0,0 +1,82 @@
+# Prompt Engineering — Exercises
+
+## Exercise 1: Improve a Vague Prompt
+
+**Task:** Someone wrote: "Make this email sound professional." Rewrite it to be specific. Include: audience, tone, length, and one example of a phrase to avoid or use.
+
+**Validation:**
+- [ ] Specifies the audience (e.g., client, internal)
+- [ ] Defines "professional" (formal? concise? friendly-but-polished?)
+- [ ] Includes length or structure (short paragraph? bullet points?)
+- [ ] Gives at least one example of desired or undesired phrasing
+
+**Hints:**
+1. "Professional" is vague — formal? concise? industry jargon?
+2. "Sound professional for a CFO" narrows audience
+3. "Use 'regarding' not 'about'; avoid exclamation points" — concrete style
+
+---
+
+## Exercise 2: Add Few-Shot Examples
+
+**Task:** You need the AI to convert product names to URL slugs. Zero-shot: "Convert to URL slug." Add 2–3 few-shot examples. Example format: "Blue Widget Pro" → "blue-widget-pro".
+
+**Validation:**
+- [ ] At least 2 examples showing input → output
+- [ ] Examples cover edge cases (capitals, spaces, special chars?)
+- [ ] Output format is consistent (lowercase, hyphens)
+
+**Hints:**
+1. Show: "Hello World" → "hello-world"
+2. Add: "Super Product 2024" → "super-product-2024" (numbers?)
+3. Consider: "O'Reilly Book" → "oreilly-book" (apostrophes)
+
+---
+
+## Exercise 3: Request Structured Output
+
+**Task:** Ask the AI to analyze a short product review and return: sentiment (positive/negative/neutral), main topic (1–3 words), and one suggested action for the business. Request JSON with keys `sentiment`, `topic`, `action`.
+
+**Validation:**
+- [ ] JSON structure is specified
+- [ ] Keys are named exactly
+- [ ] Value types are clear (string, etc.)
+- [ ] Handles edge case: ambiguous or empty input
+
+**Hints:**
+1. "Return a JSON object with keys: sentiment, topic, action"
+2. "No markdown, no explanation — JSON only"
+3. "If unclear, use 'neutral' for sentiment"
+
+---
+
+## Exercise 4: Add Chain-of-Thought
+
+**Task:** Give the AI a multi-step logic puzzle (e.g., "Three people: A, B, C. A is taller than B. B is taller than C. Who is shortest?"). First prompt: just ask for the answer. Second prompt: ask to "reason step by step, then give the answer." Compare outputs.
+
+**Validation:**
+- [ ] CoT prompt explicitly asks for reasoning first
+- [ ] User observes CoT yields more reliable answer (or catches errors)
+- [ ] User can state when CoT helps (logic, math, debugging)
+
+**Hints:**
+1. "Think through each step before answering"
+2. "Show your reasoning, then state the final answer"
+3. For complex tasks, CoT reduces confidently-wrong answers
+
+---
+
+## Exercise 5: Fix Common Mistakes
+
+**Task:** This prompt has problems: "Be creative and write something good about our product. Make it long and detailed. Use lots of adjectives." Rewrite it to avoid: vagueness, conflicting constraints, no examples. Produce a better version.
+
+**Validation:**
+- [ ] Replaces "creative" and "good" with specifics (tone, audience, format)
+- [ ] "Long" is quantified (e.g., 2 paragraphs, 150 words)
+- [ ] Adds at least one example of desired style or output
+- [ ] Removes or reconciles conflicting instructions
+
+**Hints:**
+1. "Creative" → "Convey enthusiasm without hype; avoid superlatives"
+2. "Long" → "2 paragraphs, ~100 words"
+3. Add example: "Like X, not like Y"

package/modules/prompt-engineering/game.yaml

@@ -0,0 +1,101 @@
+games:
+  - type: classify
+    title: "Prompt Technique Classifier"
+    categories:
+      - name: "Zero-Shot"
+        color: "#58a6ff"
+      - name: "Few-Shot"
+        color: "#3fb950"
+      - name: "Chain-of-Thought"
+        color: "#d29922"
+      - name: "Structured Output"
+        color: "#a371f7"
+    items:
+      - text: "Translate the following text to French: Hello, how are you?"
+        category: "Zero-Shot"
+      - text: "Here are 3 examples of sentiment: 'I love it' = positive, 'terrible' = negative, 'it's okay' = neutral. Now classify: 'Best product ever!'"
+        category: "Few-Shot"
+      - text: "Solve: If a store has 3 apples and sells half of them, then buys 5 more, how many does it have? Think step by step."
+        category: "Chain-of-Thought"
+      - text: "List the top 3 programming languages. Respond in JSON: {\"languages\": [{\"name\": \"...\", \"reason\": \"...\"}]}"
+        category: "Structured Output"
+      - text: "What is the capital of Japan?"
+        category: "Zero-Shot"
+      - text: "Example 1: 'hot' → temperature. Example 2: 'bright' → light. Example 3: 'loud' → sound. Categorize: 'fragrant'"
+        category: "Few-Shot"
+      - text: "A train leaves at 2pm going 60 mph. Another leaves at 3pm from the same station at 90 mph. When does the second catch the first? Show your reasoning."
+        category: "Chain-of-Thought"
+      - text: "Summarize this article. Output format: {\"title\": \"string\", \"keyPoints\": [\"...\"], \"wordCount\": number}"
+        category: "Structured Output"
+      - text: "Rewrite this sentence to be more professional."
+        category: "Zero-Shot"
+      - text: "Few-shot: 'run'→verb, 'quickly'→adverb, 'beautiful'→adjective. What part of speech is 'happiness'?"
+        category: "Few-Shot"
+
+  - type: speed-round
+    title: "Prompt Fix Sprint"
+    rounds:
+      - question: "The model gives inconsistent formats. What's the best fix?"
+        options:
+          - "Add more context"
+          - "Add format constraint (e.g. JSON schema, template)"
+          - "Use a different model"
+          - "Retry multiple times"
+        answer: 1
+        timeLimit: 16
+      - question: "The model fails on multi-step reasoning. What should you add?"
+        options:
+          - "More examples"
+          - "Use chain-of-thought (e.g. 'Think step by step')"
+          - "Lower temperature"
+          - "Shorter prompt"
+        answer: 1
+        timeLimit: 16
+      - question: "Outputs vary wildly between runs. What to try?"
+        options:
+          - "Add examples"
+          - "Add format constraint"
+          - "Lower temperature or set seed"
+          - "Use CoT"
+        answer: 2
+        timeLimit: 15
+      - question: "The model misunderstands edge cases. Best fix?"
+        options:
+          - "Add few-shot examples covering edge cases"
+          - "Use structured output"
+          - "Increase max tokens"
+          - "Add a disclaimer"
+        answer: 0
+        timeLimit: 16
+      - question: "Responses are too verbose. What helps?"
+        options:
+          - "Add 'Be concise' or length constraint"
+          - "Add more examples"
+          - "Use chain-of-thought"
+          - "Higher temperature"
+        answer: 0
+        timeLimit: 15
+      - question: "The model gives wrong answers on classification. Fix?"
+        options:
+          - "Add format constraint"
+          - "Add few-shot examples with correct labels"
+          - "Lower temperature only"
+          - "Use CoT"
+        answer: 1
+        timeLimit: 16
+      - question: "JSON output is malformed. What to do?"
+        options:
+          - "Add examples"
+          - "Add format constraint / schema / 'return valid JSON'"
+          - "Use CoT"
+          - "Retry on failure"
+        answer: 1
+        timeLimit: 15
+      - question: "The model skips steps in a procedure. Best approach?"
+        options:
+          - "Add format constraint"
+          - "Add explicit chain-of-thought instruction"
+          - "Lower temperature"
+          - "Add more examples"
+        answer: 1
+        timeLimit: 16

package/modules/prompt-engineering/module.yaml

@@ -0,0 +1,45 @@
+slug: prompt-engineering
+title: "Prompt Engineering — Techniques That Get Better AI Outputs"
+version: 1.0.0
+description: "Learn what makes a good prompt, zero-shot vs few-shot prompting, chain-of-thought reasoning, and structured output for better AI results."
+category: ai-and-llm
+tags: [prompt-engineering, ai, llm, claude, prompting, few-shot]
+difficulty: beginner
+
+xp:
+  read: 10
+  walkthrough: 30
+  exercise: 20
+  quiz: 15
+  quiz-perfect-bonus: 10
+  game: 20
+  game-perfect-bonus: 10
+
+time:
+  quick: 5
+  read: 15
+  guided: 45
+
+prerequisites: []
+related: [llm-fundamentals, ai-pair-programming]
+
+triggers:
+  - "How do I write better prompts?"
+  - "What is prompt engineering?"
+  - "How do I get better results from Claude?"
+  - "What are few-shot examples?"
+
+visuals:
+  diagrams: [diagram-mermaid, diagram-flow]
+  quiz-types: [quiz-drag-order, quiz-timed-choice, quiz-fill-blank]
+  game-types: [classify, speed-round]
+  playground: bash
+  slides: true
+
+sources:
+  - url: "https://docs.anthropic.com/en/docs/build-with-claude/prompt-engineering"
+    label: "Anthropic Prompt Engineering Guide"
+    type: docs
+  - url: "https://platform.openai.com/docs/guides/prompt-engineering"
+    label: "OpenAI Prompt Engineering Guide"
+    type: docs

package/modules/prompt-engineering/quick-ref.md

@@ -0,0 +1,65 @@
+# Prompt Engineering Quick Reference
+
+## Prompt Anatomy
+
+| Component | Purpose | Example |
+|-----------|---------|---------|
+| **Role** | AI persona | "You are an expert Python developer." |
+| **Context** | Background | "Codebase uses FastAPI and Pydantic." |
+| **Task** | The request | "Add validation for the email field." |
+| **Format** | Output structure | "Return JSON with keys: valid, errors." |
+| **Constraints** | Limits | "Under 100 words. No external APIs." |
+
+## Zero-Shot vs Few-Shot
+
+| Type | When to Use |
+|------|-------------|
+| **Zero-shot** | Simple, well-defined tasks |
+| **Few-shot** | Structured output, style-sensitive, edge cases |
+
+## Chain-of-Thought
+
+- Ask: "Think step by step" / "Show your reasoning"
+- Improves: logic, math, multi-step tasks
+- Reduces: confidently-wrong answers
+
+## Temperature
+
+| Range | Use Case |
+|-------|----------|
+| 0–0.3 | Factual, code, extraction, classification |
+| 0.5–0.7 | Balanced |
+| 0.7–1.0 | Creative, varied, brainstorming |
+
+## Structured Output
+
+- **JSON** — `Return a JSON object with: name, email, score`
+- **XML** — `Wrap the summary in <summary> tags`
+- **Markdown** — `Format as a markdown table with columns A, B, C`
+
+## Iteration Order
+
+1. Start simple — one clear task
+2. Add context — if output is generic
+3. Add examples — if format/style drifts
+4. Add constraints — if too long or off-topic
+5. Test edge cases — ambiguous, empty input
+
+## Common Mistakes
+
+| Mistake | Fix |
+|---------|-----|
+| Vague instructions | Be explicit: "List 3 pros and 3 cons" |
+| No examples | Add 1–2 few-shot examples |
+| Too many constraints | Add incrementally |
+| Assuming model knows context | Provide relevant background |
+| One giant prompt | Break into steps or sub-tasks |
+
+## By Task Type
+
+| Task | Strategy |
+|------|----------|
+| Summarization | Specify length, key points, audience |
+| Analysis | Define criteria, structure (pros/cons) |
+| Code | Language, framework, patterns, I/O format |
+| Creative | Tone, length, style, examples |