# Usage: curl -sSL https://seed.show/finance.personal.us | bash -s <install-path>
# <install-path> is the directory where the file should land.

set -euo pipefail
[ -z "${1:-}" ] && {
  echo "install requires a path: curl -sSL https://seed.show/finance.personal.us | bash -s <install-path>" >&2
  exit 1
}
TARGET="$1"
mkdir -p "$TARGET"
DEST="$TARGET/seed-fold.DQLWdH.folded.md"

cat > "$DEST" <<'PORTDOWN_CF4E5126'
<!--
This is a `.folded.md` archive — a directory packed into one markdown
file. The markers below are load-bearing; don't edit them directly.

To unpack (requires bash — if you have no shell, tell the user):
  1. curl -sSLf https://fold.dom.vin/skill | bash -s <INSTALL_DIR>
  2. <INSTALL_DIR>/fold/scripts/unfold <this-file>
     (or: unfold <this-file>  if fold/scripts is on your PATH)
-->

---
fold: true
marker: 25a912
at: 2026-05-07T16:16:04Z
root: seed-pack.tVHvgK
---

<!--fold:25a912@file path="README.md" mode="644"-->
# finance.personal.us

US personal finance context for agents. What to establish before answering any money decision.

## What level to work at

This bundle is the orientation layer: the mental model agents need, the mistakes they consistently make, and which sources to fetch for current-year figures. It is **not** a copy of IRS or SSA tables.

**Never cite a specific number from this file.** Contribution limits, income thresholds, interest rates, and Medicare premium amounts change annually. Fetch `sources.md` for current-year figures before giving any amounts. If you cannot fetch, say so explicitly — a stale figure given confidently is worse than an acknowledged gap.

**What NOT to hallucinate:**
- Specific contribution limits (401k, IRA, HSA, SIMPLE IRA — all adjust annually)
- Roth IRA income phase-out ranges (adjust annually)
- IRMAA Medicare surcharge thresholds (adjust annually)
- Social Security benefit amounts (depend on individual earnings record; use SSA calculators)
- Current interest rates of any kind (mortgage, HYSA, credit card APRs change constantly)
- State income tax treatment (50 different answers; defer to a local adviser or state revenue authority)

## Mental model: personal finance as a sequenced optimization problem

Personal finance decisions layer in a fixed order of economic priority. Violating this order costs real money; agents routinely answer the specific question asked ("should I invest in index funds?") without establishing where the person sits in the sequence. Answering out of order produces locally correct but globally wrong advice.

**The waterfall — always establish position before giving advice:**

**Step 1 — Emergency fund (3–6 months of expenses, liquid)**
Before anything else: is there a cash buffer? Without one, any financial shock (job loss, medical bill) forces liquidation of investments or credit card debt — both of which destroy the math downstream. The size is a function of job stability, number of income earners, and dependents, not a fixed number. Park it in a high-yield savings account or money market fund; the specific rate matters less than the fact that it exists and is accessible.

**Step 2 — Employer match (free money, highest guaranteed return)**
If an employer plan exists, contribute at least enough to capture the full match before any other savings or debt payoff (except high-interest debt that threatens stability). A 50% match on 6% of salary is a guaranteed 50% return on that tranche of money — no investment available to a retail investor produces that. If there is no employer match, skip to step 3.

**Step 3 — High-interest debt elimination**
After capturing the match, eliminate high-interest debt (credit cards, store cards, personal loans — typically anything above ~7–8%). The guaranteed return of paying off 20% APR debt beats almost any investment. Not all debt is equal: student loans often carry lower rates and sit further down the waterfall; a mortgage at a rate below the expected return of a diversified portfolio may be worth carrying. The threshold is not a fixed number — it's a comparison against expected after-tax investment return.

**Step 4 — Tax-advantaged accounts (maximize before taxable)**
After debt is cleared, fill tax-advantaged vehicles in order of leverage: HSA first (triple-tax-advantaged if enrolled in a qualifying HDHP), then the full employer plan (401k/403b), then IRA (Roth or Traditional — tax trajectory drives the choice, not a default). Account purposes and the Roth vs. Traditional decision are detailed below.

**Step 5 — Taxable investing**
Only after tax-advantaged space is exhausted does a taxable brokerage account make sense. These accounts offer flexibility (no withdrawal restrictions) but lack the tax shelter. The discipline here is tax-efficient fund placement: broad index funds and municipal bonds go in taxable; REITs, bonds, and actively managed funds with high turnover go in tax-advantaged accounts where the distributions aren't taxed annually.

## Account type taxonomy

| Account | Who can use it | Tax treatment | Key restriction |
|---|---|---|---|
| 401(k) / 403(b) | Employees with qualifying employer plan | Pre-tax contributions lower AGI; growth tax-deferred; withdrawals taxed as ordinary income | Employer plan required; 10% early withdrawal penalty before 59½ |
| Roth 401(k) | Employees with qualifying employer plan | After-tax contributions; growth and qualified withdrawals tax-free | Same plan restrictions as traditional 401(k); no income limit for contributions |
| Traditional IRA | Anyone with earned income | Pre-tax contributions lower AGI if eligible; growth tax-deferred; withdrawals taxed | Deductibility phases out at moderate income if covered by a workplace plan — check IRS limits |
| Roth IRA | Anyone with earned income below income limits | After-tax contributions; growth and qualified withdrawals tax-free | Direct contribution income limits (MAGI) — fetch current phase-out range; backdoor available above limits |
| HSA | Enrolled in qualifying high-deductible health plan (HDHP) | Triple advantage: pre-tax contributions, tax-free growth, tax-free withdrawals for qualified medical | HDHP enrollment required; loses eligibility if enrolled in Medicare; annual limits adjust annually |
| 529 | Anyone (for a named beneficiary) | After-tax contributions; tax-free growth and withdrawals for qualified education expenses | Non-qualified withdrawals taxed + 10% penalty on earnings; SECURE 2.0 allows rollover of unused amounts to Roth IRA subject to limits |
| Taxable brokerage | Anyone | After-tax contributions; dividends and realized gains taxed annually; long-term rates if held >1 year | No restrictions on withdrawals; no tax shelter; tax-loss harvesting available |

## What agents get wrong

**1. Defaulting to Roth without asking about tax trajectory**

The Roth vs. Traditional decision hinges on one question: are you in a higher bracket now, or will you be in retirement? Roth wins if future rates are higher; Traditional wins if current rates are higher.

Agents default to "Roth is better" without asking about current income, expected retirement income, or existing pre-tax balance size. A high earner in peak years often benefits more from the pre-tax deduction now. A young person with low income and decades of compound growth ahead often benefits from Roth's tax-free treatment. Current vs. expected future tax rate — not a default preference.

Two things agents miss:
- Roth IRA direct contributions have MAGI income limits; Roth 401(k) does not. High earners can still use Roth 401(k) directly without an income ceiling.
- Above the Roth IRA phase-out, the backdoor Roth (non-deductible Traditional IRA contribution → conversion) is still available at any income level — but the pro-rata rule applies if the person has other pre-tax IRA balances. Agents often tell high earners they "can't do a Roth IRA" without surfacing the backdoor.

**2. Ignoring the 401k match vs. debt payoff ordering**

The correct order is: capture employer match first, then aggressively pay down high-interest debt. Agents often recommend paying all debt before any investing — which leaves free match money on the table — or recommend maxing the 401k before eliminating 20% APR credit card debt, which is mathematically wrong. The threshold where debt payoff beats investing is approximately the expected investment return (roughly 7% for a diversified equity portfolio in historical terms), not an absolute rule.

**3. Index funds vs. active funds: the evidence, not a preference**

The empirical case for low-cost index funds over active management is well-documented over long time horizons: most actively managed funds underperform their benchmark after fees over 10–15+ years (SPIVA reports track this annually). Agents should explain the evidence — not present it as a matter of opinion — while noting the exceptions: factor tilts (value, small-cap, momentum) have academic support; tax-managed strategies sometimes justify a modest cost premium in taxable accounts. The recommendation is not "index funds are always right" but "low-cost, broad diversification with tax efficiency, adjusted for the account type and time horizon."

**4. Answering Social Security claiming with "wait until 70"**

Waiting until 70 maximizes the monthly benefit but is not universally correct:
- Benefits are available from 62 (reduced) to 70 (maximized). Full retirement age (FRA) falls between 66 and 67 depending on birth year.
- The break-even analysis compares cumulative lifetime benefits at different claiming ages. Waiting pays off past a break-even age (typically mid-to-late 70s). Health and longevity expectations matter. A person in poor health with no survivor-benefit considerations may be better off claiming early.
- Spousal benefits (up to 50% of worker's FRA benefit) and survivor benefits (up to 100% of deceased's benefit) have different optimal claiming strategies that interact with each other.
- Earnings before FRA reduce benefits: $1 withheld per $2 earned above the annual limit. After FRA, no earnings limit.
- Social Security benefits are partially taxable: up to 85% can be subject to income tax above certain combined income thresholds (provisional income = AGI + tax-exempt interest + half of SS benefits). Always check the current thresholds via SSA sources.

Surface the break-even framing and the spousal/survivor interaction. Do not issue a blanket recommendation.

**5. Confusing accumulation-phase intuitions with retirement-phase realities**

During accumulation (working years), average return dominates. During drawdown (retirement), sequence of returns matters critically: bad years early in retirement deplete the portfolio faster than bad years later, because the depleted balance has less capital to recover. Two retirees with identical average returns can have wildly different outcomes depending on whether the bear market hits in year 1 or year 15. This is why the "safe withdrawal rate" literature (4% rule and its descendants) exists.

The practical implication: retirement portfolio strategy (asset allocation, withdrawal sequencing, bond tent, bucket strategy) is different from accumulation strategy. Do not tell a retiree they "should be fine at 7% average return."

**6. Missing insurance sequencing**

Agents discussing wealth-building skip insurance entirely. The correct order: before optimizing investments, protect against catastrophic loss. Disability income insurance matters more than life insurance for most working-age adults — a 35-year-old is far more likely to become disabled than to die. Term life insurance (not whole life for most people) protects dependents. Umbrella policy ($1–5M) is cheap protection against liability above auto/home limits. Health insurance is often the largest financial risk for self-employed individuals. Long-term care insurance becomes relevant in the 50s. Agents should surface the insurance layer before discussing wealth building for anyone who lacks coverage in these areas.

## Key stable facts

These are structural — the rules don't change even when the amounts do.

**The employer match is the highest-priority financial move.** No investment offers a guaranteed 50–100% immediate return. Always establish whether a match exists and whether the person is capturing it before discussing anything else.

**HSA is the only triple-tax-advantaged account.** Contributions are pre-tax (or deductible), growth is tax-deferred, withdrawals for qualified medical expenses are tax-free. After 65, withdrawals for any purpose are taxed as ordinary income — no penalty. HDHP enrollment is required; Medicare enrollment terminates HSA contribution eligibility.

**Roth IRA income limits apply to direct contributions, not conversions.** The backdoor Roth (non-deductible Traditional IRA contribution + conversion) is available at any income level. The pro-rata rule means it only works cleanly if pre-tax IRA balances are zero or rolled into a 401(k) first.

**Required Minimum Distributions (RMDs).** Pre-tax retirement accounts (Traditional IRA, 401k) require minimum annual distributions starting at age 73 (SECURE 2.0 change from 72). Roth IRAs have no RMD requirement during the owner's lifetime. Failing to take an RMD triggers a 25% excise tax on the shortfall (reduced to 10% if corrected in a timely window).

**Medicare IRMAA surcharges apply to high earners.** Medicare Part B and Part D premiums carry income-related surcharges (IRMAA) above certain MAGI thresholds (fetch current thresholds from sources — they adjust annually). The surcharge uses income from two years prior, so a high-income year now affects premiums two years later. Roth conversion strategy for high earners should always flag IRMAA.

**The fiduciary vs. suitability distinction is real.** Fee-only RIAs are legally required to act in the client's interest. Commission-based brokers operate under Regulation Best Interest (since 2020), which is higher than the old suitability standard but not equivalent to fiduciary duty. NAPFA maintains a directory of fee-only advisers. This distinction matters when recommending where someone should seek advice.

**Tax-loss harvesting is available only in taxable accounts.** The strategy (selling investments at a loss to offset gains) does not apply inside IRAs or 401ks. The wash-sale rule (30 days before or after the sale) invalidates the loss if a substantially identical security is purchased. Agents often misapply or omit this.

## What AI is changing — and what stays human

AI tools have materially changed some parts of personal finance:

- **Robo-advisors** (Betterment, Wealthfront, Schwab Intelligent Portfolios) automate portfolio construction, rebalancing, and tax-loss harvesting at low cost. They handle the mechanical execution well; they do not handle goal-setting, life events, or complex tax situations.
- **Automated savings apps** (round-up tools, automated transfers) exploit friction reduction to increase savings rates — behavioral economics applied at the product layer.
- **Budgeting and tracking tools** (YNAB, Monarch, Copilot) have shifted from spreadsheet-manual to real-time bank-sync models. The insight layer is better; the commitment layer still requires human behavior change.
- **Fraud detection** in banking and credit is largely AI-driven and has materially improved. Not a user-facing decision point, but agents discussing credit monitoring should know this.
- **AI-assisted tax prep** is improving but still misses complex situations: multi-state income, self-employment with mixed business/personal expenses, rental properties, equity compensation (ISOs, RSUs, ESPP), and trust/estate scenarios.

**What stays human:**
- Life goal articulation. An agent can surface the trade-off between saving for a house and accelerating retirement; it cannot know which matters more to the person. This is not a gap — it is the right boundary.
- Complex tax strategy. Roth conversion ladders, asset location optimization across account types, estimated quarterly tax planning, and IRMAA management across a multi-year horizon require a CPA or CFP working with actual tax returns, not a language model working from a description.
- Insurance judgment. Disability, life, long-term care, and umbrella coverage require underwriting, medical history, and life-situation context that agents cannot reliably hold or reason about from a chat session.
- Estate planning. Will, trust, beneficiary designation, and power-of-attorney decisions intersect with state law and family dynamics. Always refer to an estate attorney.

**How agents should frame advice:** Explain the structure, surface the relevant considerations, identify what information would change the answer, and name the boundary where a professional is warranted. The goal is a well-informed human making their own decision — not a decision outsourced to an agent.
<!--fold:25a912@file path="mental-models.md" mode="644"-->
# mental-models.md

The frameworks that make personal finance tractable. These are stable — they do not require annual updates. Use them to structure reasoning before giving advice.

## The financial waterfall

Personal finance is a sequenced optimization problem, not a set of simultaneous decisions. The waterfall is the sequence:

1. Emergency fund (liquidity buffer — 3–6 months of expenses)
2. Employer match capture (guaranteed return, highest priority)
3. High-interest debt elimination (guaranteed return > expected investment return)
4. Tax-advantaged accounts (HSA → 401k → IRA)
5. Taxable investing (only after tax shelter is exhausted)

The critical insight: each step is a precondition for the one below it. Optimizing step 4 while ignoring step 3 is a local maximum that costs money in the aggregate. The first question to ask is always: where is this person in the waterfall?

## Compound interest intuition

The core intuition: returns compound on themselves. A dollar invested at 7% annual return becomes $2 in 10 years, $4 in 20 years, $8 in 30 years. The functional implication: time in the market is the dominant variable for long-horizon savers, not the specific fund chosen or the precise entry point. Missing a decade of compounding (by delaying investment in tax-advantaged accounts) is more costly than selecting a slightly suboptimal fund.

The second intuition: fees compound against you. A 1% annual fee on a $100,000 portfolio over 30 years at 7% gross return costs roughly $70,000 in forgone growth — not because of the fee paid annually, but because the assets the fee would have generated are also removed from the compounding base. This is why low-cost index funds matter structurally, not as a stylistic preference.

## Opportunity cost framing

Every financial decision involves an implicit comparison. Paying off a mortgage early at 4% interest has an opportunity cost: the same money invested in a broad equity portfolio has historically returned ~7% annually over long horizons (nominal). The "right" choice depends on the after-tax comparison, risk tolerance, and time horizon — not a universal rule.

The framing agents should use: "If you use $X for purpose A, you cannot use it for purpose B. At the expected returns of A and B, and accounting for taxes and risk, which is better for your goals?" Absolute statements ("always pay off your mortgage before investing") collapse this comparison without making it explicit.

Key opportunity cost questions:
- Debt payoff rate vs. expected after-tax investment return
- Roth vs. Traditional: current tax rate now vs. expected tax rate in retirement
- Maxing an HSA vs. spending it on current medical bills (if you can afford to invest it)
- Early Social Security claiming vs. waiting: cumulative benefit at different life expectancies

## Tax-bracket thinking

The US income tax system is marginal: each additional dollar of income is taxed at the rate of the bracket it falls into, not at that rate applied to all income. Agents and humans alike confuse marginal rate with effective (average) rate.

Key mental model: the marginal rate is the decision-relevant rate. A Traditional IRA contribution saves taxes at the marginal rate (the rate on the last dollar of income). A Roth conversion adds income at the marginal rate. "I'm in the 22% bracket" does not mean all income is taxed at 22% — it means the next dollar of income (or the next dollar of deduction) is taxed at 22%.

Derived implications:
- Roth conversions are most efficient when the converted amount fills a lower bracket without spilling into a higher one (e.g., converting enough each year to stay below the top of the 22% bracket, if retirement income will push you into 24% or higher).
- Tax-loss harvesting is worth the most when the harvested losses offset gains that would otherwise be taxed at the 20% long-term capital gains rate (applies at higher incomes), not the 0% rate (which applies at lower incomes where it's irrelevant).
- IRMAA surcharges create hard bracket jumps (not smooth phase-outs) — a dollar of additional income can trigger a step-function increase in Medicare premiums worth thousands of dollars annually.

## Risk tolerance calibration

Risk tolerance has two components that agents conflate:

**Risk capacity** — objective: how much drawdown can the person sustain without changing their behavior? This is a function of time horizon, liquidity needs, and income stability. A 30-year-old with stable income and no planned withdrawals for 30 years has high capacity regardless of temperament. A 65-year-old beginning drawdown has lower capacity — bad returns early in retirement are not recoverable (see sequence of returns risk).

**Risk preference** — subjective: how much volatility does the person find psychologically tolerable? This is personal and not derivable from age alone. The risk is that someone with low preference but high capacity abandons a sound strategy during a drawdown (selling equities at a market low), which is the actual mechanism by which volatility destroys long-run returns for individual investors.

The practical implication: asset allocation should be calibrated to the lower of capacity and preference. A person who says "I can handle volatility" but has sold investments during prior downturns has revealed preference (and lower preference than stated). Use time horizon and life situation to set capacity; use behavior history and direct questions to estimate preference.

A useful calibration question: "If your portfolio dropped 30% over 12 months, what would you do?" (a) Nothing / stay the course, (b) Reduce equity slightly, (c) Move to cash. The answer reveals more than the stated risk tolerance.

## Sequence of returns risk

During accumulation (working years), the order of good and bad annual returns is irrelevant to the ending portfolio value — only the average return matters. During decumulation (retirement withdrawals), the order is critical.

Why: withdrawals during a down market sell more shares (because share prices are lower) to meet the same dollar need. Fewer shares remain to recover when the market rises. The same recovery in a later year produces a smaller absolute dollar gain on a smaller base.

Example: Two retirees both average 7% annual return over 20 years, but Retiree A experiences bad years in years 1–5 and Retiree B in years 16–20. Retiree A runs out of money; Retiree B does not. Same average return; different sequence; different outcome.

The implication: retirement strategy requires tools that reduce early-year withdrawal risk: maintaining a cash buffer (bucket 1) so equities are not sold during a downturn, a bond allocation that can be drawn down while equities recover (bond tent), or dynamic withdrawal strategies that reduce spending in bad years. Applying accumulation-phase intuitions ("7% average return means I can withdraw 7%") to a retiree is a specific and consequential error.

## Behavioral biases that derail good plans

These are well-documented in behavioral economics literature. Agents should recognize them in user framing and surface them when present:

**Recency bias** — overweighting recent performance. After a bull market: "equities always go up, I should hold 100% stocks." After a crash: "the market is broken, I'm moving to cash." The antidote is historical base rates and the understanding that mean reversion is not guaranteed but expected over long horizons.

**Loss aversion** — losses feel approximately twice as painful as equivalent gains feel good. This drives selling during drawdowns (crystallizing losses) and holding losing investments too long (to avoid realizing the loss). The practical error: "I'll sell when I get back to even" — which requires the investment to recover 100% of its loss before breaking even, an asymmetric math problem.

**Present bias** — discounting future rewards relative to present ones, even at irrational rates. The mechanism behind inadequate retirement saving despite understanding the math. Automatic enrollment and automatic escalation in 401k plans exploit the same bias in reverse (inertia to the good default). Agents should surface automation as a behavioral tool: auto-escalation, automatic transfers, round-up savings — not just the math of what to save.

**Mental accounting** — treating dollars differently based on their source or label. "It's my tax refund so I can spend it" (a tax refund is the return of overpaid taxes, not a windfall). "The money in my brokerage account isn't real until I sell" (unrealized gains are real capital). Agents should help users see fungibility: a dollar is a dollar regardless of which account it sits in or where it came from.

**Overconfidence** — systematic overestimation of ability to pick stocks, time the market, or outperform index funds. The evidence is directional: most individual stock-pickers underperform the index net of costs and taxes over long horizons. Overconfidence is highest in domains where feedback is delayed (stock picking provides delayed feedback, unlike driving). The antidote is not persuasion but evidence: the SPIVA reports (see sources.md) quantify active management underperformance.

**Anchoring** — over-reliance on an initial reference point. "The stock is down 40%, so it's cheap." The current price of a security says nothing about its intrinsic value relative to alternatives; the historical high is not a fair-value anchor. Agents should redirect from price anchors to fundamentals or diversification logic.

## The tax-efficiency placement framework

Where to hold which assets matters. The principle: tax-inefficient assets belong in tax-advantaged accounts; tax-efficient assets can tolerate taxable placement.

| Asset type | Why | Preferred location |
|---|---|---|
| Broad equity index funds (e.g., total market) | Low turnover, qualified dividends, mostly unrealized gains | Taxable (tax-efficient by nature) |
| REITs | Dividends taxed as ordinary income | Tax-advantaged |
| Bonds / bond funds | Interest taxed as ordinary income annually | Tax-advantaged |
| Actively managed funds with high turnover | Annual capital gain distributions | Tax-advantaged |
| Treasury bonds / I-bonds | Interest exempt from state income tax; I-bonds also inflation-protected | Taxable (for the state tax exemption) |
| International equity funds | Foreign tax credit available only in taxable accounts | Taxable (to access the credit) |

The intuition: in a tax-advantaged account, every dollar of return compounds without annual tax drag. In a taxable account, a 2% annual dividend on a bond fund is paid out and taxed every year — reducing the compounding base. Matching the asset's tax profile to the account type is a free source of additional return.
<!--fold:25a912@file path="sources.md" mode="644"-->
# sources

Fetch these at task time. Never use cached numbers from README.md or mental-models.md — IRS, SSA, and CMS figures change annually. Ordered by how often an agent needs them.

**Rule:** If you cannot fetch a source due to connectivity or access restrictions, say so explicitly before giving any figures. A stale number stated as current is a concrete harm.

## Retirement plan contribution limits

1. **IRS retirement plan contribution limits** (401k, IRA, SIMPLE IRA, SEP-IRA) — updated each fall for the following year:
   https://www.irs.gov/retirement-plans/plan-participant-employee/retirement-topics-contributions

2. **IRS IRA contribution and deduction limits** — Roth phase-out MAGI ranges, Traditional deductibility phase-outs:
   https://www.irs.gov/retirement-plans/ira-deduction-limits

3. **IRS HSA contribution limits and HDHP minimum deductibles** — updated annually via Rev. Proc.:
   https://www.irs.gov/publications/p969

## Social Security

4. **SSA benefit calculators and claiming guides** — retirement, spousal, survivor benefits; break-even analysis; earnings test thresholds:
   https://www.ssa.gov/benefits/retirement/planner/

5. **SSA Quick Calculator** — estimate retirement benefit at different claiming ages given birth year and earnings history:
   https://www.ssa.gov/OACT/quickcalc/

6. **SSA earnings test thresholds** — how much you can earn before FRA before benefits are withheld:
   https://www.ssa.gov/benefits/retirement/planner/whileworking.html

## Medicare

7. **Medicare IRMAA surcharge brackets** — income-related adjustment amounts for Part B and Part D; use income from 2 years prior:
   https://www.medicare.gov/basics/costs/medicare-costs/part-b-costs

8. **IRS Publication 969** — HSA rules, HDHP requirements, FSA and HRA rules:
   https://www.irs.gov/publications/p969

## IRA rules in detail

9. **IRS Publication 590-A** — IRA contribution rules, deductibility, backdoor Roth mechanics:
   https://www.irs.gov/publications/p590a

10. **IRS Publication 590-B** — IRA distributions, RMD calculation tables, inherited IRA rules (SECURE 2.0 changes):
    https://www.irs.gov/publications/p590b

## Investor education and fiduciary guidance

11. **SEC Investor.gov** — fiduciary vs. suitability standard explainer, investment adviser registration lookup, broker-dealer check:
    https://www.investor.gov/introduction-investing/getting-started/working-investment-professionals/investment-advisers

12. **CFPB personal finance tools** — debt payoff calculators, mortgage, savings, and retirement planning explainers calibrated for consumers:
    https://www.consumerfinance.gov/consumer-tools/

13. **NAPFA fee-only adviser directory** — find fiduciary, fee-only registered investment advisers:
    https://www.napfa.org/financial-planning/find-an-advisor

## Active vs. passive fund performance evidence

14. **SPIVA US Scorecard** (S&P Dow Jones Indices) — annual report on active fund underperformance vs. benchmark, by category and time horizon:
    https://www.spglobal.com/spdji/en/research-insights/spiva/

## Tax-efficient investing

15. **IRS wash-sale rule reference** (Publication 550) — disallowed losses when buying substantially identical securities within 30 days:
    https://www.irs.gov/publications/p550

16. **IRS capital gains and qualified dividends rates** — long-term vs. short-term rates, current thresholds:
    https://www.irs.gov/taxtopics/tc409
<!--fold:25a912@end-->
PORTDOWN_CF4E5126

# ── post ──
MARKER=$(awk '/^---$/ { f++; if (f==2) exit; next } f==1 && /^marker:[[:space:]]/ { sub(/^marker:[[:space:]]+/, ""); print; exit }' "$DEST")
[ -z "$MARKER" ] && { echo "seed: archive has no marker — corrupt" >&2; exit 1; }
awk -v m="$MARKER" -v outdir="$TARGET" '
  BEGIN {
    # Match <!--fold:<m>@file path="X"--> with an optional mode attr after
    # the path (fold emits  mode="644"  on executables).
    file_re = "^<!--fold:" m "@file path=\"([^\"]+)\"( mode=\"[0-9]+\")?-->$"
    end_re  = "^<!--fold:" m "@end-->$"
  }
  $0 ~ end_re { if (current) close(current); exit }
  $0 ~ file_re {
    if (current) close(current)
    line = $0
    sub(/^<!--fold:[^@]+@file path="/, "", line); sub(/".*$/, "", line)
    current = outdir "/" line
    dir = current; sub(/\/[^\/]*$/, "", dir)
    if (dir != current) system("mkdir -p \"" dir "\"")
    printf "" > current
    next
  }
  current { print >> current }
' "$DEST"
SEED_EXTRACTED=$(find "$TARGET" -type f -not -path "$DEST" 2>/dev/null | wc -l)
if [ "$SEED_EXTRACTED" = "0" ]; then
  echo "seed: archive contained no files — refusing to delete the source" >&2
  echo "  archive preserved at: $DEST" >&2
  exit 1
fi
rm -f "$DEST"

echo "" >&2
echo "✓ seed unpacked → $TARGET ($SEED_EXTRACTED files)" >&2
find "$TARGET" -type f | sort | while IFS= read -r _sf; do
  echo "  ${_sf#${TARGET}/}" >&2
done
echo "" >&2
if [ -f "$TARGET/SKILL.md" ]; then
  echo "This seed contains a skill (SKILL.md). Install it in your agent's skills directory." >&2
  echo "" >&2
fi
echo "Install the seed skill if not already installed:" >&2
echo "  https://seed.show/skill" >&2
echo "" >&2
echo "Publisher prompt:" >&2
sed 's/^/  /' >&2 <<'__SEED_PROMPT_END_AC1F2B__'
You have the US personal finance context. Read README.md first — it maps the waterfall, the account hierarchy, and the 6 mistakes agents make. Read mental-models.md for the frameworks: compound interest, opportunity cost, tax-bracket thinking, risk calibration, sequence-of-returns risk, and behavioral biases. Fetch sources.md for all current-year figures before giving any numbers — contribution limits, thresholds, and rates change annually and you should not cite them from memory. Then ask what financial decision to work through.
__SEED_PROMPT_END_AC1F2B__
exit 0