Strategic self-audit CLI powered by AI. Four Naval Ravikant frameworks applied to your career in one session.
Most professionals never audit their leverage, knowledge, and income-generation patterns. You lack a repeatable, rigorous way to:
- Discover your rare, defensible knowledge
- Identify where you're trading time for money (instead of building equity)
- Design products that scale your expertise
- Track which income streams actually leverage you
Leverage OS fills this gap with AI-powered frameworks that synthesize Naval's thinking into actionable self-analysis.
Answer a consolidated intake questionnaire once. The CLI runs four asymmetric audit frameworks via AWS Bedrock (Claude):
| Framework | Output |
|---|---|
| Specific Knowledge Excavator | Your rare, defensible knowledge niche + differentiation thesis |
| Leverage Stack Auditor | Income streams analyzed for scale, equity, and time-for-money leaks |
| Productize Yourself Blueprint | Product/service design that leverages your knowledge without you |
| Time-for-Money Leak Detector | Conversion roadmap: rented hours → equity-based income |
All results saved as markdown for versioning and future reference.
┌─────────────────────────────────────────────────────────────┐
│ CLI Input → Consolidated Questionnaire │
│ (career, obsessions, income sources, skills) │
└────────────────┬────────────────────────────────────────────┘
│
┌────────────────▼────────────────────────────────────────────┐
│ Prompt Framework Layer │
│ ├─ SK Excavator (leverage pattern mining) │
│ ├─ Leverage Auditor (income stream analysis) │
│ ├─ Productize Blueprint (design framework) │
│ └─ Time-Leak Detector (equity roadmap) │
└────────────────┬────────────────────────────────────────────┘
│
┌────────────────▼────────────────────────────────────────────┐
│ AWS Bedrock API (Claude Model) │
│ Parallel inference across 4 frameworks │
└────────────────┬────────────────────────────────────────────┘
│
┌────────────────▼────────────────────────────────────────────┐
│ Markdown Output │
│ └─ ./outputs/{date}-leverage-os-{audit-type}.md │
└─────────────────────────────────────────────────────────────┘
Generic ChatGPT coaching prompts ask "what are your goals?" and return vague encouragement. Leverage OS applies four rigorous, interlocking frameworks derived from Naval Ravikant's wealth philosophy — each with scoring rubrics, concrete deliverables, and testable claims. The output is a structured audit you can version-control, diff annually, and act on this week — not motivational noise you forget by tomorrow.
# Install
pip install -e .
# Configure AWS credentials
aws configure
# Run full audit (all four frameworks)
leverage-os
# Or run individual audits
leverage-os --audit knowledge # Specific Knowledge Excavator only
leverage-os --audit leverage # Leverage Stack Auditor only
leverage-os --audit productize # Productize Yourself Blueprint only
leverage-os --audit time-leak # Time-for-Money Leak Detector onlyResults appear in your terminal and are automatically saved to ./outputs/ with ISO timestamps.
Below is a sample Leverage Stack Auditor result for a solo SaaS founder who also does consulting. This is the markdown that gets saved to ./outputs/.
Sample: Leverage Stack Auditor — Solo SaaS Founder
---
date: 2026-06-10
tool: leverage-os
source: aws-bedrock
tags: [naval-ravikant, leverage, self-audit]
---
# Leverage OS — Full Audit Run (2026-06-10 09:42)
## 2. The Leverage Stack Auditor
**Leverage Audit:**
| Activity | Leverage Type | Hours/Week | Score | Revenue % |
|----------|--------------|-----------|-------|-----------|
| Enterprise consulting (data strategy) | Labor | 25 | 1 | 55% |
| SaaS product (analytics dashboard) | Code | 12 | 4 | 30% |
| Technical writing (paid newsletter) | Media | 4 | 3 | 10% |
| Angel investing (2 startups) | Capital | 1 | 5 | 5% |
**Your Leverage Index:** 2.05/5
Calculation: (1 x 55 + 4 x 30 + 3 x 10 + 5 x 5) / 100 = 2.05
**Biggest Leverage Leak:** Enterprise consulting — 25 hours/week generating only 55% of revenue at a leverage score of 1. Every hour here is rented and non-compounding. You are trading your highest-value skill (data architecture) for the lowest-leverage format (live delivery). Opportunity cost: ~15 hours/week that could move your SaaS from $4.5K to $15K MRR.
**3 Upgrade Moves:**
1. **Convert "data strategy workshop" into a self-paced assessment tool** — Package your recurring client diagnostic (the first 3 sessions of every engagement) as an automated scoring workflow inside your SaaS. Score: 1 → 4. Timeline: 21 days.
2. **Turn your top 3 consulting frameworks into a gated case-study series** — Publish "How [anonymized client] reduced data pipeline costs 40% using the 3-Layer Audit" as a 5-part email sequence driving SaaS trials. Score: 1 → 3. Timeline: 14 days.
3. **Raise consulting rate 40% and cap at 15 hrs/week** — Price out low-value clients. Reinvest the freed 10 hours into SaaS feature development (the onboarding funnel you've been deferring). Score: stays 1 but hours drop, effective index rises to 2.6. Timeline: 7 days (send rate increase email Friday).
**30 Day First Move:** This week — extract the "Data Maturity Scorecard" you run in every first consulting session. Build it as a 12-question Typeform connected to a scoring spreadsheet. Gate the detailed results behind a SaaS trial signup. Ship by Friday. Named deliverable: live URL for the Data Maturity Scorecard lead magnet.
---When displayed in your terminal via Rich, each framework result renders inside a bordered panel with the title highlighted in green, and the markdown tables and headers are fully formatted.
| Component | Technology |
|---|---|
| CLI Framework | Python 3.10+ · argparse |
| LLM Backend | AWS Bedrock (Claude) |
| Interactive Input | questionary |
| Output Formatting | rich · markdown |
| Package Management | setuptools |
# Install in editable mode with dev dependencies
pip install -e ".[dev]"
# Run linting (if configured)
ruff check .
# Format code
ruff format .Create a .env file (see .env.example):
# Optional: specify AWS profile
AWS_PROFILE=your-profile
# Optional: specify AWS region (defaults to us-east-1)
AWS_REGION=us-west-2Each audit generates a timestamped markdown file:
outputs/
├── 2026-06-10-leverage-os-full-run.md # All four audits
├── 2026-06-10-leverage-os-knowledge.md # Specific Knowledge Excavator
├── 2026-06-10-leverage-os-leverage.md # Leverage Stack Auditor
├── 2026-06-10-leverage-os-productize.md # Productize Yourself Blueprint
└── 2026-06-10-leverage-os-time-leak.md # Time-for-Money Leak Detector
Each file includes:
- Framework name and purpose
- Your input summary
- AI-generated analysis
- Actionable next steps
- Timestamp for version control
Principle 1: One Session, Four Perspectives
Minimize friction. Collect input once; apply all four frameworks in parallel.
Principle 2: Rigorous Frameworks
Don't use generic prompts. Each audit implements a specific thinking model—derived from Naval's essays, decision theory, and business strategy.
Principle 3: Persistent Output
Markdown saved to disk enables versioning. Re-run annually; diff the results to track your evolution.
Principle 4: Strategic Focus
The goal is equity and leverage, not optimization. This tool helps you identify which moves actually compound.
MIT
Author: Jonathan Lyn-Shue — Fractional CIO/CTO | Data & AI Executive
Built to apply Naval Ravikant's leverage framework to modern knowledge work.