agentleFS
KameronKalesKameronKales/planfi-skills/skills/rent-vs-buy/SKILL.md

rent-vs-buy

Skill0 starsChanged 5 months agoMITView on GitHub

Decide whether to rent or buy a home by orchestrating the public planfi MCP. Use whenever someone asks "should I rent or buy", "is it cheaper to buy a $X house vs renting at $Y/mo", "rent vs buy breakeven", "when does buying beat renting", or wants to compare owning a home against renting + investing the difference over time — with the headline being the real home-appreciation rate at which buying ties renting.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 4 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    All math lives server-side — the simulation runs in nominal dollars internally (so a fixed-rate mortgage P&I is correctly held flat in …

    line 10
  2. Very specific:

    Use whichever name your environment exposes (bare or mcp__planfi__-prefixed).

    line 24
  3. Very specific:

    Call analyze_rent_vs_buy with the gathered inputs:

    line 67
  4. Very specific:

    If you already built a plan, pass { plan_id } instead of magi + filing_status and let the server derive the household context (it reports …

    line 89
  5. Distinctive:

    If the user already owns their home and is equity-rich (or a retiree weighing liquidity / decumulation), the rent-vs-buy framing often …

    line 113
  6. Distinctive:

    All decimals are fractions.

    line 154
  7. Distinctive:

    Not financial advice. Planning estimates only (approximate 2026 brackets/limits).

    line 161

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Rent vs Buy

A thin orchestration layer over the planfi MCP (https://ai.planfi.app/mcp/free). All math lives server-side — the simulation runs in nominal dollars internally (so a fixed-rate mortgage P&I is correctly held flat in nominal terms and erodes in real terms), then deflates every output back to today's (real) dollars. This skill only gathers inputs and calls the tools — it does not compute anything locally. Read-only; it never changes the user's data.

Step 0 — Make sure the planfi tools are connected

This skill uses these tools (may be namespaced, e.g. mcp__planfi__analyze_rent_vs_buy):

  • analyze_rent_vs_buy (PRIMARY) — the full rent-vs-buy engine.
  • optional: analyze_property_return (return on a specific property purchase), forecast_str_market (if they're weighing a short-term rental), generate_financial_plan (to derive household tax context and get a plan_id).
  • analyze_home_equity — model tapping existing equity (HELOC revolving draw/repay, cash-out refi, or reverse mortgage) instead of selling.

Use whichever name your environment exposes (bare or mcp__planfi__-prefixed); below they are written bare. If they're NOT available, tell the user to connect the MCP, then continue:

claude mcp add --transport http planfi https://ai.planfi.app/mcp/free

Try free, then add your key. The command above adds the free connector — https://ai.planfi.app/mcp/free (no key needed). Once you create an API key, add a new connector with the MCP url — https://ai.planfi.app/mcp — and authorize it with your key.

(On claude.ai: add a custom connector pointing at https://ai.planfi.app/mcp/free.)

Access — free for personal use. The planfi MCP is free to try (a small monthly allowance, no key needed). Heavy automated abuse forced us to add limits — but it stays free for personal use: email kam@rateapi.dev and we'll send you a free API key, no charge. (Companies and commercial use have paid plans.) To use a key, pass it as an Authorization: Bearer pft_… header in your MCP client config.

Step 1 — Gather inputs

Core scenario:

  • Home: purchase_price, down_payment_percent, closing_costs_percent, mortgage_rate (fixed nominal annual), mortgage_term_years, selling_cost_pct.
  • Carrying costs (tool assembles these into the all-in non-P&I owning cost): property_tax_rate, maintenance_pct, hoa_monthly, renters_insurance_annual.
  • Rent side: monthly_rent (today's $), annual_rent_growth_real.
  • Rates (all REAL annual, as fractions): home_appreciation_real, index_return_real (market return for invest-the-difference), and the single explicit inflation_rate that grosses the real rates up to nominal.
  • Already own it? years_already_owned, plus optional current_balance and current_home_value overrides.
  • Horizons: horizons array of years (default [5,10,20,30]).

Household tax context (REQUIRED to derive the capital-gains rate): EITHER a plan_id (from generate_financial_plan — plan-derived MAGI + filing status TAKE PRECEDENCE), OR both magi and filing_status (single / married_joint) directly. If neither is supplied the tool returns MISSING_HOUSEHOLD_CONTEXT — ask for them.

Engine facts to bake in: all dollars are today's (real) dollars; all decimals are fractions (7% → 0.07); appreciation / rent-growth / market rates are REAL; the sim runs in nominal then deflates; the fixed mortgage P&I is nominal (erodes in real terms); the §121 home-sale exclusion ($500k MFJ / $250k single) is modeled nominal (not inflation-indexed, so it erodes in real terms over long horizons); tax is federal only in v1 (LTCG 0/15/20 + 3.8% NIIT, 2026 brackets).

Step 2 — Run the analysis

Call analyze_rent_vs_buy with the gathered inputs:

analyze_rent_vs_buy({
  purchase_price: 525000,
  down_payment_percent: 0.2,
  mortgage_rate: 0.065,
  mortgage_term_years: 30,
  property_tax_rate: 0.011,
  maintenance_pct: 0.01,
  monthly_rent: 2600,
  annual_rent_growth_real: 0.01,
  home_appreciation_real: 0.015,
  index_return_real: 0.07,
  inflation_rate: 0.03,
  horizons: [5, 10, 20, 30],
  magi: 150000,
  filing_status: "single"
})
// already own it? add: years_already_owned, current_home_value, current_balance

If you already built a plan, pass { plan_id } instead of magi + filing_status and let the server derive the household context (it reports magi_source: "plan").

Step 3 — Present the results

  • Lead with the headline — headline.breakeven_home_appreciation_real (BOTH pre_tax and after_tax): the real home-appreciation rate at which BUY net worth ties RENT net worth at the longest horizon. Then the verdict string ("At your X% real appreciation assumption, RENTING/BUYING is ahead by $Y at N years.").
  • Per-horizon table (per_horizon[]): for each horizon show BUY vs RENT net worth (pre- and after-tax, all REAL dollars), the winner, and the delta (buy − rent).
  • Timeline: payoff_year (when the mortgage hits zero) and crossover_year (first year owning's monthly cost drops below rent, after P&I falls off — or null).
  • Opportunity cost: opportunity_cost block (buy vs rent terminal wealth, multiples, winner).
  • Tax breakdown: tax_breakdown — filing status, MAGI + magi_source, renter gain/tax/NIIT, home gain, §121 exclusion, home taxable gain/tax/NIIT, and the horizon it's reported at.
  • Risk flags: surface risk.flags (break-even is sensitive to the appreciation assumption; §121 exclusion erodes in real terms).
  • Surface assumed_defaults[] — read back each assumed field + assumed_value + note so the user can correct any silent assumption (these also ride on disclosures.assumptions). The server is the source of truth for what it defaulted; don't enumerate defaults from memory.

Step 4 — Already own? Consider tapping equity instead of selling

If the user already owns their home and is equity-rich (or a retiree weighing liquidity / decumulation), the rent-vs-buy framing often isn't the real question — the question is whether to tap the existing equity rather than sell. Route to analyze_home_equity to model the three levers: a HELOC (revolving interest-only draw, then amortizing repay), a cash-out refinance (new 1st lien, break-even on the payment delta), or a reverse mortgage (HECM principal limit, tenure / lump-sum / line-of-credit payout, crossover age). It returns an afterTaxBorrowingCost, an extract / do_not_extract / toss_up recommendation, and an asModelDebt for the forecast.

analyze_home_equity({
  mode: "heloc",
  homeValue: 600000,
  mortgageBalance: 250000,
  extractAmount: 80000,
  proceedsUse: "home_improvement",
  itemizesDeductions: true,
  marginalTaxRate: 0.32,
  filingStatus: "married_joint"
})  // mode: "cash_out_refi" or "reverse_mortgage" for the other levers

Next-actions hints

The tool returns a next_actions[] array (each { tool, why, prefilled_args }, with prefilled_args carrying { plan_id } when available). The only server-defined edge out of analyze_rent_vs_buy is to generate_financial_plan (fold the rent-or-buy decision into the full FIRE timeline) — follow that rather than guessing the next call. (analyze_property_return chains the other way: it suggests analyze_rent_vs_buy as ITS follow-up, not vice versa.)

Related skills

  • long-term-rental-analyzer — after-tax buy-and-hold rental returns (Schedule E P&L, 27.5-yr depreciation shelter, §1250 recapture, NIIT, 1031 deferral) when the home is an investment rental rather than a primary residence.
  • str-investment-analyzer — nightly short-term rental (Airbnb/VRBO) investment analysis.
  • home-equity (analyze_home_equity) — if they already own, route here to model HELOC / cash-out refi / reverse mortgage equity extraction instead of selling (and this tool routes back to rent-vs-buy when the question is really own-vs-rent).

Notes

  • All decimals are fractions; all dollars are today's (real) dollars; the sim runs nominal internally then deflates; rates are real.
  • The fixed mortgage P&I is nominal and the §121 exclusion is nominal — both erode in real terms, which is precisely why the real break-even appreciation rate is the right headline.
  • Tax is federal only in v1 (no state tax); LTCG 0/15/20 + 3.8% NIIT on 2026 brackets.
  • Pass plan_id to derive household MAGI + filing status; caller magi + filing_status are the fallback when there's no plan.
  • Not financial advice. Planning estimates only (approximate 2026 brackets/limits).

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.