A repeatable intake-and-diagnosis SOP that gathers the financial context people usually leave out, then grades the resulting advice against life-cycle finance principles. Education, not regulated advice.
Run a structured personal-finance review: collect the full financial picture (including the context people routinely omit), produce spending/saving/investing guidance grounded in life-cycle principles, then self-grade that guidance against a fixed set of diagnostic tests. This is educational analysis, not regulated financial advice.
When you want a structured, repeatable financial review from AI instead of a one-off chat — it collects the context you would forget and checks the advice against life-cycle principles.
Gather my complete financial picture. Ask for anything missing before advising — do NOT proceed on a partial picture, because the omitted context (not the model) is what produces generic advice.
State the baseline assumptions you will use, and flag every gap. For each missing number, say what you will assume and how sensitive the advice is to it. Use ~2% real return for safe savings and the long-run historical average for diversified stocks; never assume market timing or above-market returns.
Produce the core guidance: a spending/saving split for working years, an age-appropriate asset allocation with a glide path that reduces equity as I age, an explicit rebalancing rule (calendar + threshold), account priority (match first, then tax-advantaged, then taxable), and a high-interest-debt plan. Describe any product generically and disclose it is one option, not an endorsement.
Self-grade the guidance you just gave against the diagnostic tests below (these are the life-cycle checks the research proposes). For each, mark PASS or FIX and, if FIX, correct the guidance before finishing. Do not skip a test to look decisive.
Hand off responsibly. Summarize what to do this week, what would change the recommendation, and exactly what to verify with a licensed professional. Reinforce that this is education, not regulated advice.
Return five sections: (1) Intake summary — the picture you collected and what is still missing. (2) Assumptions — each with its biggest risk. (3) The plan — savings rate, allocation + glide path, rebalancing rule, account priority, debt plan. (4) Diagnostic scorecard — the five tests above, each PASS/FIX, with any fixes already applied. (5) Handoff — this week's actions, what would change the advice, and what to verify with a licensed professional. Keep it concrete and grounded only in my stated numbers.
Copy this skill into your library to inject it into Claude, ChatGPT, and Gemini — or install your whole library as /korvai: commands in Claude Code.
Get started free →Related skills
Turn a raw sales call transcript into a polished follow-up email and structured CRM notes in minutes.
Break a large user story into small, independently shippable stories with clear acceptance criteria.
Transform raw work logs into a polished, stakeholder-ready weekly status report in minutes.