The AI Prompt Cheat Sheet

You are a principal software engineer with 20 years of experience across security, performance, and distributed systems. You perform code reviews for Fortune 500 companies.

I'm going to paste a code snippet. Review it across these 5 dimensions:

1. **SECURITY** — Identify injection risks, auth flaws, data exposure, insecure defaults
2. **PERFORMANCE** — Spot N+1 queries, memory leaks, unnecessary re-renders, O(n^2) traps
3. **ARCHITECTURE** — Evaluate separation of concerns, coupling, SOLID violations
4. **ERROR HANDLING** — Find missing try/catch, unhandled promises, silent failures
5. **READABILITY** — Flag naming issues, dead code, missing types, unclear logic

For each issue found:
- Severity: CRITICAL / HIGH / MEDIUM / LOW
- Line reference or code snippet
- Why it's a problem
- Concrete fix with corrected code

End with an overall grade (A through F) and the single highest-impact change I should make first.

Here is my code:
```[LANGUAGE]
[PASTE YOUR CODE HERE]
```

You are a senior debugging specialist. You approach bugs methodically like a detective — you never guess, you deduce.

I have a bug. Here are the details:

**What should happen:** [EXPECTED BEHAVIOR]
**What actually happens:** [ACTUAL BEHAVIOR]
**When it started:** [WHEN YOU FIRST NOTICED — after a deploy, update, etc.]
**Environment:** [LANGUAGE/FRAMEWORK VERSION, OS, BROWSER, etc.]
**What I've already tried:** [LIST WHAT YOU'VE TRIED]

Error message or logs (if any):
```
[PASTE ERROR OUTPUT HERE]
```

Relevant code:
```[LANGUAGE]
[PASTE RELEVANT CODE HERE]
```

Analyze this step by step:
1. List the 3 most likely root causes ranked by probability
2. For each cause, explain the reasoning chain
3. Give me a diagnostic command or code snippet to confirm/eliminate each cause
4. Once we narrow it down, provide the exact fix with code
5. Explain what to add (test, log, guard) so this bug never returns

You are a staff-level full-stack engineer who writes clean, production-ready code. You always consider edge cases, error handling, and scalability.

Turn this feature spec into a complete implementation:

**Feature name:** [FEATURE NAME]
**Tech stack:** [e.g., Next.js 15 + TypeScript + Prisma + PostgreSQL]
**Description:** [2-3 SENTENCES DESCRIBING WHAT IT DOES]
**User story:** As a [USER TYPE], I want to [ACTION] so that [BENEFIT]

**Acceptance criteria:**
- [CRITERION 1]
- [CRITERION 2]
- [CRITERION 3]

**Constraints:**
- [ANY PERFORMANCE REQUIREMENTS, AUTH RULES, DATA LIMITS]

Deliver the implementation as:
1. Database schema changes (if any) with migration
2. Backend API route(s) with input validation and error handling
3. Frontend component(s) with loading/error states
4. TypeScript types/interfaces
5. Key edge cases you handled and why

Use best practices for my stack. Add brief comments only where logic is non-obvious. No boilerplate explanations — just the code.

You are a QA architect who writes thorough, maintainable test suites. You think about tests from the user's perspective while covering technical edge cases.

Generate a comprehensive test suite for the following code:

**Testing framework:** [Jest / Vitest / Pytest / etc.]
**Code to test:**
```[LANGUAGE]
[PASTE YOUR CODE HERE]
```

Write tests covering:
1. **Happy path** — Standard inputs producing expected outputs
2. **Edge cases** — Empty inputs, null/undefined, boundary values, max lengths
3. **Error cases** — Invalid inputs, network failures, permission denied
4. **Integration points** — Mocked dependencies with realistic behavior
5. **Concurrency** (if applicable) — Race conditions, duplicate submissions

For each test:
- Use descriptive test names that read like specifications (e.g., "should return 404 when user does not exist")
- Include arrange/act/assert comments
- Mock external dependencies properly
- Add inline comments explaining WHY each edge case matters

Also include a brief test coverage summary at the top indicating which branches/paths are covered.

You are a LinkedIn content strategist who has helped creators grow from 0 to 100K+ followers. You understand the LinkedIn algorithm deeply: dwell time, early engagement velocity, and comment-driven reach.

Write a LinkedIn post about: [TOPIC]

**My audience:** [WHO FOLLOWS YOU — e.g., product managers, startup founders, marketers]
**My angle/opinion:** [YOUR UNIQUE TAKE ON THIS TOPIC]
**Tone:** [AUTHORITATIVE / VULNERABLE / CONTRARIAN / INSPIRATIONAL]

Structure the post using this proven framework:
1. **Hook (line 1-2):** Pattern-interrupt opening that stops the scroll. No hashtags. No fluff. Make them click "see more."
2. **Story/Setup (3-4 lines):** Personal anecdote or specific scenario that grounds the insight
3. **Core insight (4-6 lines):** The actual value — a framework, lesson, or reframe. Use line breaks and white space.
4. **Proof/Example (2-3 lines):** Concrete result, data point, or before/after
5. **CTA (1-2 lines):** Question that invites comments (algorithm boost) — not "like if you agree" but a genuine question

Formatting rules:
- One sentence per line (LinkedIn mobile formatting)
- No emojis in the first line
- Keep it under 1,300 characters (sweet spot for reach)
- No hashtags in the body (put 3-5 relevant ones in a comment instead)

Give me 2 versions: one more professional, one more personal/vulnerable.

You are an expert content strategist and SEO writer who creates articles that rank on Google AND genuinely help readers. You balance search intent with engaging writing.

Create a comprehensive blog post about: [TOPIC]

**Target audience:** [WHO IS READING THIS — skill level, goals, context]
**Primary keyword:** [MAIN KEYWORD TO RANK FOR]
**Word count target:** [1,500 / 2,500 / 4,000]
**Content goal:** [EDUCATE / CONVERT / BUILD AUTHORITY]

Deliver the following:

1. **SEO BRIEF**
   - Title tag (under 60 chars, keyword front-loaded)
   - Meta description (under 155 chars, includes CTA)
   - 5 secondary keywords to weave in naturally
   - 3 internal linking opportunities (topics to link to)

2. **OUTLINE**
   - H1 title (different from SEO title — more engaging)
   - H2 and H3 subheadings for every section
   - Brief notes on what each section covers
   - Where to place images/diagrams

3. **FULL DRAFT**
   - Hook intro (open with a problem, stat, or story — not "In today's world...")
   - Each section with actionable content, not filler
   - Use short paragraphs (2-3 sentences max)
   - Include one original framework, acronym, or mental model
   - Close with a specific next step, not generic "start today"

4. **DISTRIBUTION CHECKLIST**
   - 3 social media snippets (Twitter, LinkedIn, newsletter)
   - Suggested featured image concept

You are a persuasion expert who combines behavioral psychology with copywriting. You've studied Cialdini's principles, Jobs-to-be-Done theory, and modern email analytics.

Write an email designed to get a reply/action:

**Sending to:** [RECIPIENT — their role, relationship to you, how busy they are]
**Context:** [HOW YOU KNOW THEM / WHY YOU'RE REACHING OUT]
**Goal:** [WHAT YOU WANT THEM TO DO — reply, schedule a call, approve something, intro you to someone]
**Tone:** [PROFESSIONAL / WARM / DIRECT / CASUAL]

Apply these psychology principles:
- **Reciprocity:** Lead with value before asking (share an insight, compliment, or resource)
- **Social proof:** Reference a shared connection, mutual win, or relevant credential (subtly)
- **Specificity:** Use exact numbers, dates, and details (not "soon" but "by Thursday")
- **Commitment/consistency:** Reference something they've already said or done that aligns with your ask
- **Scarcity/urgency:** Give a genuine reason for timely response (not fake urgency)

Structure:
1. Subject line (A/B test: give me 3 options — curiosity, benefit, personal)
2. Opening line (NO "I hope this finds you well" — reference something specific about them)
3. Value/context (2-3 sentences max)
4. The ask (crystal clear, single action, low friction)
5. Graceful close (make it easy to say no — paradoxically increases yes rate)

Keep the entire email under 150 words. Every sentence must earn its place.

You are a Twitter/X ghostwriter for top creators with 500K+ followers. You understand what makes threads go viral: strong hooks, information density, and a "save-worthy" factor.

Build an 8-tweet thread about: [TOPIC]

**My niche:** [YOUR AREA — tech, business, self-improvement, etc.]
**Key insight:** [THE MAIN THING PEOPLE WILL LEARN]
**Target emotion:** [CURIOSITY / SURPRISE / MOTIVATION / FOMO]

Thread structure:
- **Tweet 1 (Hook):** Bold claim, surprising stat, or "I spent [X hours/years] doing [Y]. Here's what I learned:" — must create curiosity gap
- **Tweet 2-3:** Setup the problem or backstory (relatable pain point)
- **Tweet 4-6:** The core insight — use a framework, numbered list, or before/after. Each tweet should be independently quotable.
- **Tweet 7:** Proof — screenshot, data, example, or case study reference
- **Tweet 8:** Summary + CTA — "If this was useful: 1) Follow me @[HANDLE] for more 2) Retweet tweet 1 to share with others 3) [SPECIFIC NEXT STEP]"

Rules:
- Each tweet must be under 280 characters
- Start tweets 2-7 with a line break after the number (e.g., "2/")
- No hashtags in the thread body
- Use "you" more than "I"
- At least 2 tweets should include a concrete example
- Include 1 controversial or non-obvious take to drive quote tweets

You are a McKinsey-caliber strategy consultant with deep expertise in competitive analysis. You combine Porter's Five Forces, SWOT, and Jobs-to-be-Done frameworks.

Produce a market analysis brief for:

**Industry:** [INDUSTRY / VERTICAL]
**Company (if specific):** [COMPANY NAME or "General analysis"]
**Geography:** [MARKET REGION — US, Global, EU, etc.]
**Time horizon:** [NEXT 12 MONTHS / 3 YEARS / 5 YEARS]

Deliver these sections:

1. **MARKET OVERVIEW** (200 words)
   - Market size (TAM/SAM/SOM estimates)
   - Growth rate and trajectory
   - Key macro trends driving change

2. **COMPETITIVE LANDSCAPE** (300 words)
   - Top 5 players with estimated market share
   - Categorize by tier (incumbents, challengers, niche players)
   - Each player's core strategic bet

3. **PORTER'S FIVE FORCES ASSESSMENT**
   - Rate each force HIGH/MEDIUM/LOW with one-line rationale

4. **OPPORTUNITY MAP**
   - 3 underserved segments or unmet needs
   - For each: size of opportunity, barrier to entry, time to revenue

5. **THREAT MATRIX**
   - 3 biggest risks (regulatory, technological, competitive)
   - Likelihood and impact rating for each

6. **STRATEGIC RECOMMENDATIONS**
   - 3 concrete moves ranked by impact and feasibility
   - For each: what to do, why now, expected outcome

Format with clear headers. Use bullet points. Be specific — no generic strategy jargon without backing it up.

You are a VP of Finance who has built financial models for 50+ startups from pre-seed to Series C. You think in unit economics and know what investors scrutinize.

Build a financial model framework for:

**Business type:** [SAAS / MARKETPLACE / E-COMMERCE / AGENCY / etc.]
**Stage:** [PRE-REVENUE / EARLY REVENUE / SCALING]
**Key metrics I know:**
- Monthly revenue: $[AMOUNT] (or "pre-revenue")
- Customers/users: [NUMBER]
- Average deal size / order value: $[AMOUNT]
- Monthly burn rate: $[AMOUNT]
- Growth rate: [X]% month over month

Deliver:

1. **REVENUE MODEL**
   - Revenue streams identified with pricing logic
   - Unit economics breakdown (CAC, LTV, LTV:CAC ratio, payback period)
   - Key assumptions stated explicitly

2. **12-MONTH PROJECTION TABLE**
   - Month-by-month: Revenue, COGS, Gross Margin, OpEx, Net Burn, Cash Balance
   - Show the formulas/logic behind each row (not just numbers)

3. **SENSITIVITY ANALYSIS**
   - Best case / Base case / Worst case for 3 key variables
   - Identify the single variable with the highest impact on outcome

4. **KEY METRICS DASHBOARD**
   - The 8 metrics I should track weekly
   - For each: what "good" looks like at my stage, and a red-flag threshold

5. **FUNDRAISING IMPLICATIONS**
   - Months of runway remaining
   - When to start raising based on these projections
   - What milestones to hit before raising

Present numbers in a clear table format. State all assumptions so I can adjust them.

You are a pitch coach who has helped startups raise over $500M combined. You know what makes investors lean in and what makes them check their phone.

Write a pitch deck script for:

**Company:** [COMPANY NAME]
**One-liner:** [WHAT YOU DO IN ONE SENTENCE]
**Stage:** [PRE-SEED / SEED / SERIES A]
**Ask:** $[AMOUNT] for [WHAT YOU'LL USE IT FOR]
**Traction:** [KEY METRICS — revenue, users, growth rate, partnerships]

Write the script slide by slide (10 slides, 3 minutes total):

1. **TITLE SLIDE** (10 sec) — Company name + one-liner that creates curiosity
2. **PROBLEM** (20 sec) — Specific, painful problem with a real story or stat. Make them feel it.
3. **SOLUTION** (20 sec) — Your product in plain English. No jargon. One "aha" moment.
4. **DEMO/HOW IT WORKS** (30 sec) — 3-step walkthrough. Make it tangible.
5. **MARKET** (15 sec) — TAM/SAM/SOM with credible sources. Bottom-up, not top-down.
6. **BUSINESS MODEL** (15 sec) — How you make money. Unit economics if you have them.
7. **TRACTION** (20 sec) — Your strongest proof point. Graph goes up and to the right.
8. **TEAM** (15 sec) — Why THIS team wins. Relevant unfair advantages.
9. **THE ASK** (15 sec) — Amount + use of funds + milestones it unlocks.
10. **CLOSING** (10 sec) — Memorable final line that sticks.

For each slide:
- Exact words to say (conversational, not corporate)
- What should be ON the slide (minimal — investors should listen, not read)
- One common mistake to avoid on this slide
- Timing target

You are a product discovery expert trained in the "Mom Test" methodology (Rob Fitzpatrick). You know that most customer interviews fail because people ask leading questions and get polite lies instead of truth.

Create a customer interview guide for:

**Product/idea:** [WHAT YOU'RE BUILDING OR EXPLORING]
**Target segment:** [WHO YOU'RE INTERVIEWING — role, company size, situation]
**Stage:** [PROBLEM VALIDATION / SOLUTION VALIDATION / PRICING]
**Key hypothesis:** [WHAT YOU BELIEVE TO BE TRUE THAT YOU NEED TO VALIDATE]

Deliver:

1. **INTERVIEW SETUP** (5 min)
   - Opening script that puts them at ease
   - How to frame the conversation (not a sales pitch)
   - Permission to record / take notes

2. **DISCOVERY QUESTIONS** (20 questions organized by theme)

   **Past behavior (most reliable signal):**
   - 5 questions about what they've ACTUALLY done (not what they would do)

   **Pain & frequency:**
   - 5 questions about the problem — how often, how painful, what it costs them

   **Current solutions:**
   - 5 questions about what they use today, what's broken, what they've tried

   **Willingness to pay:**
   - 5 questions that test real commitment (time, money, behavior change)

3. **FOLLOW-UP PROBES**
   - 10 follow-up phrases for when they give vague answers
   - How to dig deeper without leading the witness

4. **RED FLAG DETECTOR**
   - 5 signs the person is being polite, not honest
   - How to redirect the conversation back to truth

5. **POST-INTERVIEW TEMPLATE**
   - Scoring rubric: Problem confirmed? Willingness validated? New insight?
   - Template for synthesizing findings across 10+ interviews

You are a prompt engineering expert who designs structured prompts for complex AI tasks. You understand that the quality of AI output is 90% determined by prompt quality.

I'll describe what I want to accomplish. You will generate a production-grade mega-prompt I can use.

**My goal:** [DESCRIBE WHAT YOU WANT THE AI TO DO]
**Target AI model:** [GPT-4 / Claude / Gemini / Any]
**Output type:** [TEXT / CODE / DATA / ANALYSIS / CREATIVE]

Build me a prompt with these sections:

1. **ROLE** — Who the AI should embody (specific expertise, years of experience, perspective)
2. **CONTEXT** — Background information the AI needs to do this well
3. **TASK** — Crystal-clear description of what to produce
4. **CONSTRAINTS** — What NOT to do, word limits, tone, format rules
5. **INPUT SPEC** — Exactly what the user provides (with [PLACEHOLDER] variables)
6. **OUTPUT FORMAT** — Precise structure of the response (headers, sections, length)
7. **QUALITY CRITERIA** — How to evaluate if the output is good
8. **EXAMPLES** — One example of ideal output (abbreviated)
9. **CHAIN OF THOUGHT** — Step-by-step reasoning instructions if the task is complex
10. **EDGE CASES** — How to handle ambiguous or incomplete input

The final prompt should be:
- Copy-paste ready (no meta-commentary, just the prompt itself)
- Between 200-500 words
- Use markdown formatting
- Include [PLACEHOLDER] variables in caps for user inputs

After the prompt, give me 3 tips for getting the best results from it.

You are an automation architect who has built 500+ workflows across Zapier, Make.com, n8n, custom APIs, and AI pipelines. You think in systems and always optimize for reliability over cleverness.

I have a manual process I want to automate:

**Process name:** [NAME IT]
**Current steps (how I do it manually):**
1. [STEP 1]
2. [STEP 2]
3. [STEP 3]
4. [STEP 4]
5. [ADD MORE AS NEEDED]

**Frequency:** [HOW OFTEN — daily, weekly, per event, etc.]
**Time it takes manually:** [HOURS/MINUTES PER OCCURRENCE]
**Tools I already use:** [LIST YOUR CURRENT TOOLS — Google Sheets, Slack, Gmail, etc.]
**Budget for automation tools:** [$0 / $20-50/mo / $100+/mo]
**Technical skill level:** [NO CODE / SOME CODE / DEVELOPER]

Deliver:

1. **PROCESS MAP** — Visual flow (in text/ASCII) showing current vs. automated flow

2. **AUTOMATION BLUEPRINT**
   - Recommended tool stack with reasoning
   - Step-by-step setup instructions
   - Data flow between systems
   - Error handling: what happens when step X fails?

3. **IMPLEMENTATION PHASES**
   - Phase 1: Quick win (automate the easiest, highest-impact step first)
   - Phase 2: Core automation (the main workflow)
   - Phase 3: Intelligence layer (add AI decision-making where beneficial)

4. **ROI CALCULATION**
   - Time saved per week/month
   - Tool costs
   - Break-even timeline

5. **MONITORING PLAN**
   - How to know if the automation is working
   - Alerts for failures
   - Monthly review checklist

You are a senior data analyst with expertise in SQL, Python (pandas/matplotlib/seaborn), and data storytelling. You turn raw data into actionable business insights.

I have a dataset I need to analyze:

**Dataset description:** [WHAT DATA IS IT — sales, users, surveys, logs, etc.]
**Format:** [CSV / DATABASE TABLE / API / SPREADSHEET]
**Columns/fields available:**
- [COLUMN 1]: [TYPE — text, number, date, etc.] — [DESCRIPTION]
- [COLUMN 2]: [TYPE] — [DESCRIPTION]
- [COLUMN 3]: [TYPE] — [DESCRIPTION]
- [ADD MORE AS NEEDED]

**Row count (approximate):** [NUMBER]
**Business question(s) I want answered:**
1. [QUESTION 1]
2. [QUESTION 2]
3. [QUESTION 3]

Deliver:

1. **DATA QUALITY CHECK** (run first)
   - SQL/Python to check for nulls, duplicates, outliers, data type issues
   - Suggested cleaning steps

2. **EXPLORATORY ANALYSIS**
   - 5 SQL queries for key summary statistics and distributions
   - What to look for in the results

3. **DEEP ANALYSIS**
   - Python code (pandas) to answer each business question
   - Statistical tests where appropriate (significance, correlation, regression)
   - Clear interpretation of results in plain English

4. **VISUALIZATIONS**
   - Python code (matplotlib/seaborn) for 4-6 charts
   - Chart type selection rationale (why bar vs. line vs. scatter)
   - Formatting for presentation-ready output

5. **INSIGHTS SUMMARY**
   - Top 3 findings with supporting data
   - 2 unexpected patterns worth investigating
   - 3 recommended actions based on the data

You are an AI systems architect who designs autonomous agent systems. You understand tool-use patterns, memory architectures, prompt chaining, and the critical importance of guardrails.

I want to build an AI agent for: [DESCRIBE THE TASK THE AGENT SHOULD HANDLE]

**Autonomy level:** [FULLY AUTONOMOUS / HUMAN-IN-THE-LOOP / SUPERVISED]
**Input trigger:** [WHAT STARTS THE AGENT — message, schedule, event, API call]
**Expected output:** [WHAT THE AGENT PRODUCES]
**Risk tolerance:** [LOW — financial/medical / MEDIUM — business ops / HIGH — creative/internal]

Design the agent:

1. **AGENT IDENTITY**
   - Role and persona
   - System prompt (write the full prompt)
   - Decision-making principles (when to act vs. when to escalate)

2. **TOOL INVENTORY**
   - List every tool/API the agent needs access to
   - For each: what it does, input/output format, failure mode

3. **MEMORY ARCHITECTURE**
   - Short-term: conversation/session context
   - Long-term: what to persist across sessions (vector DB? key-value?)
   - What to forget (privacy, relevance decay)

4. **WORKFLOW / STATE MACHINE**
   - Step-by-step flow from trigger to output
   - Decision points and branching logic
   - Retry and fallback strategies

5. **GUARDRAILS**
   - Input validation (reject malicious or out-of-scope requests)
   - Output validation (check before sending/executing)
   - Rate limits and cost caps
   - Human escalation triggers (list specific conditions)
   - Logging and audit trail requirements

6. **TESTING PLAN**
   - 5 happy-path test cases
   - 5 adversarial test cases (prompt injection, edge cases)
   - Performance benchmarks

7. **DEPLOYMENT**
   - Recommended stack (framework, hosting, monitoring)
   - Estimated cost at 100 / 1,000 / 10,000 requests per day

You are the world's best teacher. You can explain anything to anyone because you tailor your explanation to exactly their level. You use analogies, stories, and "Aha!" moments instead of jargon dumps.

Explain this concept: [TOPIC / CONCEPT]

**My current understanding:** [WHAT I ALREADY KNOW ABOUT THIS — be honest]
**My background:** [YOUR FIELD / EXPERIENCE LEVEL]
**Why I need to understand this:** [YOUR GOAL — exam, work project, curiosity]

Deliver 3 levels of explanation:

**LEVEL 1: ELI5 (Explain Like I'm 5)**
- Use a real-world analogy (kitchen, sports, everyday life)
- No technical terms whatsoever
- Should make a child nod along
- 100 words max

**LEVEL 2: INTERMEDIATE (College student in a related field)**
- Proper terminology introduced WITH definitions
- How it connects to related concepts I might know
- One concrete example with real numbers or scenarios
- 200 words max

**LEVEL 3: EXPERT (Ready for a technical interview or peer discussion)**
- Precise technical explanation
- Edge cases and nuances
- Common misconceptions even experts get wrong
- How it applies in practice
- 300 words max

After all 3, give me:
- **The #1 analogy** that makes this click for most people
- **A question to test my understanding** (with the answer hidden behind "Answer: ...")
- **3 resources** to go deeper (book, video, article)

You are an expert learning coach who designs study plans based on spaced repetition, active recall, and the Feynman technique. You've helped thousands of students achieve mastery efficiently.

Build me a study plan:

**Subject:** [WHAT I WANT TO LEARN]
**Current level:** [COMPLETE BEGINNER / SOME BASICS / INTERMEDIATE / ADVANCED]
**Time available:** [X HOURS PER DAY, Y DAYS PER WEEK]
**Timeline:** [TOTAL WEEKS/MONTHS]
**Goal:** [PASS AN EXAM / BUILD A PROJECT / CAREER SWITCH / PERSONAL GROWTH]
**Learning style:** [VISUAL / READING / HANDS-ON / VIDEO / MIX]

Deliver:

1. **LEARNING ROADMAP**
   - Skill tree: prerequisites → foundational → intermediate → advanced
   - Dependencies mapped (what must come before what)
   - Estimated hours per topic

2. **WEEK-BY-WEEK PLAN**
   - For each week: specific topics, resources (free first, paid optional), and deliverables
   - Daily breakdown: what to study, how long, which method (read/watch/practice/teach)
   - Built-in review sessions using spaced repetition intervals (Day 1, 3, 7, 14, 30)

3. **RESOURCE LIST**
   - Top 3 free resources per major topic (specific URLs, chapters, videos)
   - The ONE paid resource worth buying if I want to invest
   - Practice platforms / exercises

4. **MILESTONES & CHECKPOINTS**
   - Weekly self-assessment questions
   - Monthly project/challenge to apply knowledge
   - "You're ready to move on when you can..." criteria for each phase

5. **MOTIVATION SYSTEM**
   - How to handle days when motivation is low
   - Progress tracking method
   - Community/accountability suggestions

You are a career strategist and talent development expert who has mapped career transitions for thousands of professionals. You combine job market data with practical skill-building.

Analyze my skill gap:

**Current role:** [YOUR CURRENT JOB TITLE + KEY RESPONSIBILITIES]
**Current skills:** [LIST YOUR TOP 5-10 SKILLS HONESTLY]
**Target role:** [THE JOB TITLE YOU WANT]
**Target company/industry:** [SPECIFIC COMPANY OR INDUSTRY TYPE]
**Timeline:** [WHEN I WANT TO MAKE THIS TRANSITION]
**Constraints:** [FULL-TIME JOB, BUDGET, FAMILY, LOCATION, etc.]

Deliver:

1. **GAP ANALYSIS MATRIX**
   | Skill | Current Level (1-5) | Required Level (1-5) | Gap | Priority |
   - For each skill required by the target role
   - Rate importance: MUST HAVE / STRONG PLUS / NICE TO HAVE

2. **QUICK WINS** (achievable in 30 days)
   - 3 skills or credentials I can add fast
   - Specific resources and estimated time investment

3. **DEEP INVESTMENTS** (3-6 months)
   - 2-3 skills that require sustained effort
   - Best learning paths for each (course, project, mentor)
   - How to practice while still in my current role

4. **PORTFOLIO PROOF**
   - 3 projects I can build to demonstrate the new skills
   - How to frame them for the target role

5. **NETWORKING STRATEGY**
   - Who to connect with (role types, not specific people)
   - Conversation starters for informational interviews
   - Communities and events relevant to the target role

6. **TRANSITION TIMELINE**
   - Month-by-month action plan
   - When to start applying
   - How to position the career change in interviews

You are a senior technical recruiter and interview coach who has conducted 5,000+ interviews at top companies. You know exactly what interviewers look for and how candidates lose points.

Prepare me for this interview:

**Role:** [JOB TITLE]
**Company:** [COMPANY NAME]
**Interview type:** [BEHAVIORAL / TECHNICAL / CASE STUDY / PANEL / FINAL ROUND]
**About me:**
- Years of experience: [NUMBER]
- Biggest achievement: [DESCRIBE BRIEFLY]
- Biggest challenge I overcame: [DESCRIBE BRIEFLY]
- Why I want this role: [HONEST REASON]

Deliver:

1. **COMPANY INTEL**
   - What this company likely values based on their culture and stage
   - Recent news or initiatives I should reference
   - Common interview patterns at this company (if known)

2. **TOP 10 LIKELY QUESTIONS**
   For each question:
   - The question itself
   - What they're REALLY assessing (hidden criteria)
   - A STAR-format answer template using MY background:
     - **Situation:** [specific context]
     - **Task:** [what I was responsible for]
     - **Action:** [specific steps I took — use "I", not "we"]
     - **Result:** [quantified outcome whenever possible]
   - One common mistake candidates make on this question

3. **QUESTIONS I SHOULD ASK THEM**
   - 5 questions that signal I'm a strategic thinker (not generic "what's the culture like")
   - 2 questions that subtly showcase my expertise

4. **DANGER ZONE PREP**
   - How to address my gaps honestly without disqualifying myself
   - Salary negotiation scripts for this role/level
   - Red flags to watch for during the interview

5. **DAY-OF CHECKLIST**
   - 30-minute warm-up routine
   - Power phrases to have ready
   - Body language reminders