@muskanmeet/invoicing-design-system 0.1.0

package/.local/secondary_skills/insurance-optimizer/SKILL.md

@@ -0,0 +1,209 @@
+---
+name: insurance-optimizer
+description: Review insurance coverage and find opportunities to optimize premiums and reduce gaps.
+---
+
+# Insurance Optimizer
+
+Review current insurance coverage, identify gaps or overpayment, and suggest strategies to optimize premiums. Covers auto, home/renters, health, and life insurance.
+
+**DISCLAIMER: This provides general information only, not professional insurance or financial advice.**
+
+## When to Use
+
+- User wants to review if they're over- or under-insured
+- User is paying too much and wants to save
+- User wants to understand their coverage
+- User is shopping for new insurance
+
+## When NOT to Use
+
+- Filing insurance claims
+- Complex commercial insurance
+- Specific policy interpretation (consult agent)
+
+## Methodology
+
+### Step 1: Gather Current Coverage
+
+Ask the user for their current policies. For each, collect:
+
+- **Type**: Auto, home/renters, health, life, umbrella, disability
+- **Provider**: Who is the carrier?
+- **Premium**: Monthly or annual cost
+- **Deductible**: How much they pay before insurance kicks in
+- **Coverage limits**: Maximum the policy will pay
+- **Key features**: What's included, what's excluded
+
+### Step 2: Assess Coverage by Type
+
+#### Auto Insurance
+
+**Minimum recommended coverage:**
+
+| Coverage | Recommended Minimum | Notes |
+|----------|-------------------|-------|
+| Bodily injury liability | 100/300 ($100K per person, $300K per accident) | State minimums are dangerously low |
+| Property damage liability | $100,000 | Covers damage to other vehicles/property |
+| Uninsured/underinsured motorist | Match liability limits | Protects you from uninsured drivers |
+| Collision | Based on car value | Consider dropping if car value < 10× annual premium |
+| Comprehensive | Based on car value | Covers theft, weather, animals |
+| Medical payments / PIP | $5,000-$10,000 | Covers your medical costs regardless of fault |
+
+**Drop collision/comp test:** If car value (KBB private party) < 10× the annual collision+comp premium, OR < ~$4,000 outright → self-insure. A $3,000 car with a $1,000 deductible and $400/yr premium means best-case payout is $2,000 — you're paying 20%/yr to insure that.
+
+**Liability floor:** 100/300/100 minimum. State minimums (often 25/50/25) won't cover a single hospital visit. If net worth >$500k, bump to 250/500/100 + umbrella.
+
+#### Home / Renters Insurance
+
+**Homeowners:**
+
+| Coverage | Guideline |
+|----------|-----------|
+| Dwelling | Full replacement cost (NOT market value) |
+| Personal property | Enough to replace belongings (do a home inventory) |
+| Liability | $300,000-$500,000 minimum |
+| Additional living expenses | 20% of dwelling coverage |
+| Deductible | $1,000-$2,500 (higher = lower premium) |
+
+**Commonly missed:** Flood (NOT in standard policies — check FEMA zone), earthquake (separate), scheduled riders for jewelry/art >$1,500-2,500, sewer backup, home business equipment.
+
+**Renters:** $15-30/mo for $30-50k property + $100-300k liability. Best value in insurance — never skip.
+
+#### Health Insurance
+
+| | HMO | PPO | HDHP + HSA |
+|---|---|---|---|
+| Premium | Lower | Higher | Lowest |
+| Deductible | Lower | Moderate | Highest ($1,650+ ind.) |
+| Network | Referral needed | Any | Any |
+| Best for | Low utilization | Frequent specialists | Healthy + want tax savings |
+
+**HSA advantage:** Triple tax benefit — contributions deductible, growth tax-free, withdrawals tax-free for medical. 2026 limits: $4,400 individual / $8,750 family (+$1,000 if 55+). The only account in the tax code better than a Roth IRA. Save receipts — reimburse yourself decades later, tax-free.
+
+**HDHP break-even math:** `(PPO premium − HDHP premium) × 12 + employer HSA contribution` = your buffer. If expected annual healthcare spending < buffer + tax savings on HSA contribution, HDHP wins. Most healthy people under 50 without chronic conditions come out ahead on HDHP.
+
+#### Life Insurance
+
+**How much:**
+
+- Rule of thumb: 10-12× annual income
+- More precise: Calculate total financial obligations (mortgage, debts, children's education, income replacement for X years) minus existing assets
+
+**Term vs. Whole — run the math:**
+
+| | Term (20yr, $500k) | Whole ($500k) |
+|---|---|---|
+| Monthly, healthy 30yo | ~$25-30 | ~$200-450 (8-15× term) |
+| Cash value after 20yr | $0 | ~$50-80k (2-4% IRR after fees) |
+| "Buy term, invest the difference" | $300/mo in index fund @ 7% real → **~$150k** after 20yr | — |
+
+The salesperson's commission on whole life is typically 50-100% of the first-year premium — that's why it's pushed hard. Whole life makes sense only for: estate tax planning above the ~$15M exemption, special-needs trust funding, or maxed-out every other tax-advantaged account and still have excess.
+
+**Term life is right for 90%+ of people.** Ladder policies (e.g., $500k/30yr + $500k/20yr) to match declining need as mortgage shrinks and kids age out.
+
+#### Umbrella Insurance
+
+**The $500k trigger:** Standard auto/home liability maxes out at ~$300-500k. Once your attachable net worth (home equity + taxable brokerage + savings — exclude 401k/IRA, they're federally protected from most judgments) crosses ~$500k, you're a lawsuit target without a shield.
+
+- Coverage = total attachable net worth, rounded up to nearest $1M
+- Cost: **~$150-300/yr for first $1M**, each additional $1M only ~$75-100/yr. $5M runs ~$500-700/yr. The cheapest insurance per dollar of coverage in existence.
+- **Prerequisite:** most carriers require $250-300k underlying liability on auto/home before writing umbrella
+- **Get it if:** net worth >$500k, rental properties, teenage drivers, pool/trampoline/dog, coach youth sports, high public profile, or you post opinions on the internet under your real name
+
+### Step 3: Identify Gaps
+
+Common coverage gaps to flag:
+
+- [ ] Liability limits too low relative to net worth
+- [ ] No umbrella policy
+- [ ] No disability insurance (protects income — most overlooked insurance)
+- [ ] No flood/earthquake in at-risk area
+- [ ] Renters without renters insurance
+- [ ] Life insurance insufficient for dependents
+- [ ] Health plan doesn't cover needed specialists
+- [ ] No scheduled coverage for high-value items
+
+### Step 4: Find Savings
+
+**Deductible break-even math (compute this, don't guess):**
+
+```text
+break_even_years = (high_deductible − low_deductible) / annual_premium_savings
+```
+
+- $500 → $1,000 deductible typically saves 15-30% on collision/comp (NOT proportional — doubling deductible does not halve premium)
+- Avg driver files a claim every **6-8 years**. If break-even < 3 years and you have the emergency fund, raise it.
+- Example: $500→$1,000 saves $200/yr → break-even 2.5yr → clearly worth it. Saves only $50/yr → 10yr break-even → skip.
+- **Bank the savings** in a dedicated account until it equals your highest deductible — self-insure the gap.
+
+**Shopping cadence — loyalty is a tax:**
+
+- **Auto: re-quote every 6 months** (standard policy term). Carriers use "price optimization" — they raise rates on customers their models predict won't shop. ~22% of shoppers who compare find a cheaper rate. Early-shopper discounts: up to 10-15% for quoting before your current policy expires.
+- **Home: re-quote every 2-3 years** or after any claim-free stretch
+- **Life: re-quote after health improvements** — quit smoking 12+ months, lost significant weight, A1C normalized. Rates can drop 50%+.
+- **Trigger events** that should always prompt a re-quote: birthday (esp. 25), violation falls off record (~3yr), credit score jump, marriage, move, paid off car
+
+**Savings strategies by impact:**
+
+| Strategy | Savings | Notes |
+|---|---|---|
+| Shop every 6mo (auto) | 15-30% | The Zebra, Insurify, or independent agent — get 3+ quotes |
+| Raise deductibles | 10-25% | Only if emergency fund covers it; do break-even math |
+| Bundle home + auto | 10-25% | But quote unbundled too — bundle discount sometimes masks one overpriced policy |
+| Drop collision/comp | 100% of that premium | When car value < ~10× annual premium OR < $4,000 |
+| Pay annually | 5-10% | Avoids monthly installment fees |
+| Telematics (Progressive Snapshot etc.) | 10-30% for safe drivers | Can also RAISE rates — know your driving |
+| Credit score improvement | 5-25% | Insurers use credit-based insurance scores in most states |
+
+**Comparison sites:** The Zebra / Insurify (auto+home), Policygenius (life+disability), Healthcare.gov (ACA). All free. An independent broker who writes for multiple carriers beats a captive agent (State Farm/Allstate only sell their own).
+
+### Step 5: Prioritize — Gaps Before Savings
+
+Fix underinsurance first (existential risk), then optimize premiums (efficiency).
+
+## Output Format
+
+```text
+# Insurance Review: [Name]
+
+## Current Coverage Summary
+| Type | Provider | Premium | Deductible | Coverage | Assessment |
+|------|----------|---------|-----------|----------|------------|
+| Auto | [co] | $X/mo | $Y | 100/300/100 | Adequate |
+| Home | [co] | $X/mo | $Y | $Z dwelling | Gap: flood |
+| ... | | | | | |
+
+## Total Annual Cost: $X,XXX
+
+## Gaps Identified
+1. **[Gap]** — [risk explanation and recommendation]
+
+## Savings Opportunities
+1. **[Strategy]** — estimated savings: $X-Y/year
+
+## Action Items
+1. [ ] [Highest priority action]
+2. [ ] [Next priority]
+3. [ ] [Shop for quotes by date]
+
+## Disclaimer
+General information only. Consult a licensed insurance professional for specific policy advice.
+```
+
+## Best Practices
+
+1. **Review annually** — needs, rates, and life circumstances change
+2. **Shop around** — get 3+ quotes; loyalty rarely gets the best rate
+3. **Understand deductibles** — ensure your emergency fund can cover them
+4. **Don't underinsure to save** — $50/month savings isn't worth major exposure
+5. **Ask about discounts** — most insurers have unadvertised discounts (ask explicitly)
+6. **Read the exclusions** — know what's NOT covered, not just what is
+
+## Limitations
+
+- Cannot provide actual quotes or bind policies
+- Cannot compare specific policy documents (recommend an independent agent for that)
+- Cannot interpret specific policy language or coverage disputes
+- Not a licensed insurance advisor
+- Rates and regulations vary significantly by state

package/.local/secondary_skills/interview-prep/.fingerprint

@@ -0,0 +1 @@
+ba23823f76b4a24193c5f8902b2579c4

package/.local/secondary_skills/interview-prep/SKILL.md

@@ -0,0 +1,349 @@
+---
+name: interview-prep
+description: Build mock interview simulators with voice, case interviews, behavioral prep, and scorecards.
+---
+
+# Interview Prep Simulator
+
+Instructions for building and improving AI-powered mock interview simulators that adapt dynamically to any company, role, industry, and market based on user input.
+
+## Core Principle: Dynamic Adaptation
+
+The simulator must NEVER be hardcoded to a specific company or market. Instead:
+
+1. The user provides their **target company**, **role/position**, and **location/market** during the pre-interview setup
+2. The AI interviewer uses this context to dynamically research and adapt: pulling in relevant company facts, industry dynamics, regional economic context, and role-specific technical questions
+
+3. The system prompt instructs the AI to act as an informed interviewer at that specific company and tailor all questions, scenarios, and feedback accordingly
+
+This means a single simulator can prep someone for a PE Principal role at Goldman Sachs in New York, a consulting Associate at McKinsey in London, or a VP of Finance at a regional bank in Santo Domingo — all driven by what the user enters.
+
+## Pre-Interview Setup Screen
+
+Before starting any interview, show a setup screen collecting:
+
+### Required Inputs
+
+1. **Company Name** — Text input with placeholder (e.g., "Goldman Sachs", "Banco Popular Dominicano")
+2. **Role / Position** — Text input (e.g., "Private Equity Principal", "Senior Consultant", "VP of Finance")
+
+3. **Interview Type Selector** — Card-based selection:
+
+- **Structured Interview** (icon: Briefcase) — Behavioral + technical + firm knowledge, 8–10 questions
+- **Consulting Case Interview** (icon: BarChart3) — Business case scenarios with quantitative analysis
+
+- **Behavioral Only** (icon: MessageSquare) — Focused STAR-method practice, 8–10 behavioral questions
+
+Each card shows title, brief description, estimated duration, and question count
+
+1. **Language Selector** — Dropdown or pill-toggle:
+
+- English (default), Spanish, French, Portuguese, German, Mandarin, Japanese, Arabic, Hindi
+- Show language name in both English and native script (e.g., "Spanish — Español")
+
+- Any language the AI model supports should be available
+
+### Optional Inputs
+
+1. **Industry** — Dropdown: Finance/Banking, Consulting, Technology, Healthcare, Energy, Real Estate, Other (with text input)
+2. **Difficulty Level**:
+
+- Standard: Constructive feedback, moderate follow-ups
+- Challenging: Aggressive follow-ups, stress-test answers, shorter patience for vague responses
+
+1. **Focus Areas** (checkboxes, structured interview only):
+
+- Behavioral/STAR, Technical, Deal/Project Experience, Firm & Market Knowledge, Culture Fit
+
+1. **Additional Context** — Optional textarea for the user to paste a job description, specific topics to focus on, or personal background the AI should consider
+2. **"Begin Interview" CTA** — Prominent button at the bottom; disabled until company + role + type are filled
+
+## Supported Interview Types
+
+### 1. Structured Interview (Default)
+
+Adapts question categories to the role and industry:
+
+#### For finance/PE/banking roles
+
+- Behavioral / STAR (2–3 questions)
+- Technical (LBO, valuation, capital structure, accounting) (2–3 questions)
+
+- Deal / Transaction Experience (1–2 questions)
+- Firm & Market Knowledge (1–2 questions)
+
+- Culture Fit (1 question)
+
+##### For consulting roles
+
+- Behavioral / STAR (2–3 questions)
+- Problem Solving / Frameworks (2–3 questions)
+
+- Client Experience / Engagement Stories (1–2 questions)
+- Firm & Market Knowledge (1–2 questions)
+
+- Culture Fit (1 question)
+
+###### For technology roles
+
+- Behavioral / STAR (2–3 questions)
+- System Design / Architecture (2–3 questions)
+
+- Past Projects / Technical Impact (1–2 questions)
+- Company & Product Knowledge (1–2 questions)
+
+- Culture Fit (1 question)
+
+###### For general / other roles
+
+- Behavioral / STAR (2–3 questions)
+- Role-Specific Technical (2–3 questions)
+
+- Experience & Accomplishments (1–2 questions)
+- Company Knowledge (1–2 questions)
+
+- Culture Fit (1 question)
+
+### 2. Consulting Case Interview
+
+Business-case-style interviews with quantitative and qualitative analysis:
+
+- **Market Sizing**: Top-down / bottom-up estimation relevant to the target company's industry
+- **Profitability Analysis**: Revenue/cost decomposition, margin drivers
+
+- **Market Entry**: Go/no-go framework, competitive landscape, regulatory considerations
+- **M&A / Due Diligence**: Synergy analysis, integration risk, valuation
+
+- **Operations Optimization**: Process improvement, capacity planning, cost reduction
+
+The AI selects a case scenario relevant to the target company and industry. For example:
+
+- Banking company → "Should [Company] enter the digital payments market in [Region]?"
+- Tech company → "A client's SaaS platform is losing enterprise customers — diagnose and recommend"
+
+- Healthcare → "Evaluate the acquisition of a regional hospital chain"
+
+Case flow: Scenario presentation → Clarifying questions → Framework building → Quantitative analysis → Recommendation → Evaluation
+
+### 3. Behavioral-Only Interview
+
+Focused STAR storytelling practice:
+
+- 8–10 behavioral questions across: leadership, teamwork, failure/resilience, initiative, conflict resolution, influence without authority, ambiguity, time pressure
+- Strict STAR-method feedback after every answer
+
+- Scoring on: specificity, quantification, personal ownership ("I" vs "we"), structure, and relevance to the target role
+
+## Multi-Language Support
+
+### Implementation Rules
+
+- Present the language selector on the setup screen before starting the session
+- The system prompt must include an explicit language instruction at the TOP: `"Conduct this entire interview in [language_name]. All questions, feedback, and the final scorecard must be in [language_name]."`
+
+- The UI chrome (buttons, labels, sidebar) remains in English unless the user explicitly requests full localization
+- The AI should use professional, business-appropriate register in the selected language — not casual or overly academic
+
+- For non-English interviews, the AI should still understand if the candidate mixes in English technical terms (e.g., "LBO", "IRR", "EBITDA") without penalizing them
+
+## System Prompt Architecture
+
+### Dynamic System Prompt Construction
+
+Build the system prompt dynamically from the user's setup selections. The frontend constructs the full prompt and passes it to the backend via the `systemPrompt` field on conversation creation.
+
+### System Prompt Template
+
+```text
+
+[LANGUAGE INSTRUCTION — if non-English]
+
+You are a senior interviewer at {company_name} conducting a {interview_type} interview for the {role_name} position.
+
+COMPANY CONTEXT:
+
+Research and incorporate what you know about {company_name}:
+
+- Industry position, key products/services, competitive advantages
+- Recent news, strategic initiatives, financial performance
+
+- Market/region: {location_context}
+- Company culture, values, and what they look for in candidates
+
+Use this knowledge to make questions specific and relevant. If the candidate mentions something about the company, validate or challenge their knowledge.
+
+{INTERVIEW TYPE SPECIFIC INSTRUCTIONS}
+
+INTERVIEW GUIDELINES:
+
+- Ask ONE question at a time
+- After each answer, provide brief constructive feedback (3–5 sentences max):
+
+* For behavioral: STAR structure quality, specificity, quantification, ownership ("I" vs "we")
+
+* For technical: accuracy, logical flow, assumptions stated
+
+* For cases: framework quality, math accuracy, creativity, communication
+
+- Rate each answer: Strong / Adequate / Needs Improvement
+- Then ask the next question
+
+- Be professional, direct, and constructive
+- {difficulty_instruction}
+
+FINAL SCORECARD:
+
+After all questions are complete, provide a final scorecard with:
+
+- Overall rating (Strong Hire / Hire / Lean Hire / No Hire)
+- Category-by-category scores
+
+- Top 3 strengths observed
+- Top 3 areas for improvement
+
+- Specific recommendations for interview day at {company_name}
+
+Start by briefly introducing yourself as the interviewer at {company_name}, explaining the interview format, and asking the first question.
+
+```
+
+### Difficulty Instructions
+
+- **Standard**: `"Be supportive and constructive. Give the candidate time to think. Provide helpful feedback."`
+- **Challenging**: `"Be demanding. Push back on vague answers. Ask pointed follow-ups. Challenge assumptions. Simulate a high-pressure interview environment."`
+
+## Response Timer
+
+Add a visible timer in the chat area:
+
+- Starts counting when the AI finishes asking a question (streaming ends)
+- Displays elapsed time next to the input area (e.g., "Response time: 1:32")
+
+- Stops when the user submits their answer
+- Records per-question response times for the final scorecard
+
+- Visual cue: Green < 2 min, Yellow 2–4 min, Red > 4 min
+- Timer helps candidates practice pacing — real interviews penalize overly long or short answers
+
+## End-of-Session Scorecard
+
+When the AI sends the final scorecard, detect it and render a special scorecard UI:
+
+- Parse the scorecard from the AI's markdown response
+- Display as a styled card with:
+
+- Overall rating (color-coded: green for Strong Hire/Hire, yellow for Lean Hire, red for No Hire)
+- Category-by-category scores in a visual grid
+
+- Top 3 strengths (green checkmarks)
+- Top 3 areas for improvement (amber indicators)
+
+- Response time summary (average, fastest, slowest)
+- Specific recommendations for interview day
+
+## Interview Progress Sidebar
+
+The right sidebar dynamically reflects the interview type and adapts labels to the role:
+
+**Structured Interview stages** (adapt labels to role/industry):
+
+- Finance: Behavioral → Technical/LBO → Deal Experience → Firm Knowledge → Culture Fit
+- Consulting: Behavioral → Problem Solving → Client Experience → Firm Knowledge → Culture Fit
+
+- Tech: Behavioral → System Design → Past Projects → Company Knowledge → Culture Fit
+- General: Behavioral → Technical → Experience → Company Knowledge → Culture Fit
+
+### Consulting Case stages
+
+1. Scenario Presentation → 2. Clarifying Questions → 3. Framework → 4. Analysis → 5. Recommendation
+
+#### Behavioral stages
+
+1. Leadership → 2. Teamwork → 3. Conflict → 4. Initiative → 5. Failure/Growth
+
+Each stage shows: number badge, label, active/complete/upcoming state, and a contextual tip for the current stage.
+
+## Technical Architecture
+
+### Backend (API Server)
+
+- `POST /api/openai/conversations`accepts optional`systemPrompt`,`interviewType`, and`language` fields
+- If `systemPrompt` is provided, use it instead of any default; otherwise fall back to a generic interview prompt
+
+- The streaming endpoint (`POST /api/openai/conversations/:id/messages`) remains unchanged — reads all messages including system message from DB
+- Model: use the latest available model; `max_completion_tokens: 8192`
+
+### Frontend (React + Vite)
+
+- Setup screen is the default view (no conversation ID yet)
+- On "Begin Interview," construct the dynamic system prompt from user inputs, create conversation, then start streaming
+
+- The system prompt message is hidden from the chat display — filter by `m.role === "system"` or content-matching
+- Use `cn()`from`@/lib/utils` for all dynamic classNames — avoid template literals in JSX className props (known design-subagent bug pattern)
+
+- SSE parsing: `fetch`+`ReadableStream`reader; split chunks on`\n`, parse`data: {...}` lines
+- Use `react-markdown`+`remark-gfm` for rendering AI responses
+
+### Styling
+
+- Use a professional, corporate design system — clean typography, muted palette, subtle shadows
+- Card-based setup screen with clear visual hierarchy
+
+- Adapt accent colors if desired, but default to a neutral professional palette
+
+## File Structure Reference
+
+```text
+
+artifacts/mock-interview/src/
+
+├── pages/
+
+│ └── Interview.tsx \# Main interview page (setup + chat)
+
+├── components/
+
+│ ├── Layout.tsx \# App shell with sidebar navigation
+
+│ ├── SetupScreen.tsx \# Pre-interview setup form
+
+│ ├── ChatArea.tsx \# Message list + input + timer
+
+│ ├── ProgressSidebar.tsx \# Interview stage tracker
+
+│ └── Scorecard.tsx \# Final scorecard renderer
+
+├── lib/
+
+│ ├── prompts.ts \# System prompt builder (takes setup inputs, returns prompt string)
+
+│ └── utils.ts \# cn() and helpers
+
+└── App.tsx
+
+artifacts/api-server/src/routes/openai/index.ts \# Streaming endpoint
+
+lib/api-spec/openapi.yaml \# API contract
+
+lib/db/src/schema/index.ts \# DB schema
+
+```
+
+## Improvement Checklist
+
+When improving the simulator, prioritize in this order:
+
+1. [ ] Pre-interview setup screen (company, role, type, language inputs)
+2. [ ] Dynamic system prompt builder from user inputs
+
+3. [ ] Consulting case interview support
+4. [ ] Behavioral-only interview mode
+
+5. [ ] Response timer
+6. [ ] End-of-session scorecard UI
+
+7. [ ] Difficulty level selector
+8. [ ] Focus area filtering (structured interview only)
+
+9. [ ] Session history / past interview review
+10. [ ] PDF export of scorecard and transcript

package/.local/secondary_skills/invoice-generator/.fingerprint

@@ -0,0 +1 @@
+2d6453c3413be4486e4a810b31fb1e10