Back to Discover

fjarmalalaesi

skill

jokull

Agent skills for Icelandic home finances — verðtrygging modeling, séreignarsparnaður, staðgreiðsla tax math (fjármálalæsi)

View on GitHub
14 starsMITSynced Aug 2, 2026

Install to Claude Code

/plugin marketplace add jokull/fjarmalalaesi

README

fjármálalæsi

An agent skill for Icelandic personal-finance literacy — payslips and wage rights, the three-pillar pension system, verðtryggð (CPI-indexed) mortgages, savings and investments, taxes, and safety nets. Built so anyone can point an AI agent at their own numbers and get correct, Iceland-specific answers instead of generic financial-calculator output.

Fjármálalæsi ("financial literacy") is what Icelanders call the skill of understanding money. Iceland's system is genuinely idiosyncratic — mortgage principals grow with inflation, ~90% of workers are covered by kjarasamningar that set wage floors and benefit funds, pensions come in three layers with wildly different tax and means-testing treatment, and tax on extra income depends entirely on what it stacks on. Generic tools get all of it wrong.

What the skill covers

  • Payslips & wages: launaseðill line-by-line math, union benefits and rights (kjarasamningar), salary benchmarking for negotiation
  • Pensions: the three pillars, séreignarsparnaður trade-offs (including the tax-free mortgage redirect reinstated June 2026), fund/provider choice
  • Housing: correct verðtrygging mechanics validated against real bank data, prepayment and snowball valuation in today's kronur, the mortgage market and its regulation
  • Taxes: staðgreiðsla bracket stacking — the number that decides most "should I take the money now or later" questions
  • Institutions & savings: banks vs neobanks, deposit insurance, funds, government bonds, investment taxation
  • Safety nets & life events: fæðingarorlof, atvinnuleysisbætur, námslán, barnabætur, debt help, renting
  • Data sources: verified APIs (Hagstofa PX-Web, Arion calculator) and official sources with their lag traps documented

Scripts are Python 3 stdlib-only — nothing to install.

Install

Claude Code (plugin marketplace):

/plugin marketplace add jokull/fjarmalalaesi
/plugin install fjarmalalaesi@fjarmalalaesi

Any agent supporting the Agent Skills standard:

npx skills add jokull/fjarmalalaesi

Manual copy:

git clone https://github.com/jokull/fjarmalalaesi
cp -r fjarmalalaesi/skills/fjarmalalaesi ~/.claude/skills/

Try it standalone

The scripts work without any agent:

cd skills/fjarmalalaesi/scripts

# What does a 5M prepayment on a verðtryggt loan actually buy, in today's kronur?
python3 verdtrygging.py lumpsum --balance 76000000 --apr 4.49 --months 276 \
    --inflation 4.3 --amount 5000000

# Effective tax on a séreign withdrawal stacked on a 600k/mo pension
python3 tax.py stacked --withdrawal 480000 --other-income 600000

# Live CPI and the current verðtrygging index from Hagstofa
python3 cpi.py latest

# The whole paydown-vs-invest decision in one command
python3 paydown_vs_invest.py --loans 76000000:4.49:276,8800000:5.99:276 \
    --pot 30000000 --tax-rate-now 0.4629 --inflation 4.3 --fund-return 7.0 \
    --horizon-years 20 --retirement-other-income 300000 --drawdown-years 15

Caveats

Decision-support modeling, not financial advice. Tax brackets, benefit amounts, and program rules are point-in-time (2026) — the scripts and references say where to verify current values (skatturinn.is, sedlabanki.is, hms.is, island.is). Verify big moves with your bank, union, or a licensed advisor.

License

MIT

Rendered live from jokull/fjarmalalaesi's GitHub README — not stored, always reflects the source repo.

1 Plugin

NameDescriptionCategorySource
fjarmalalaesiIcelandic personal-finance literacy: payslip and wage-rights math, the three-pillar pension system, verðtryggð mortgage modeling, séreignarsparnaður trade-offs, savings landscape, stacked staðgreiðsla tax math, and safety nets — with validated stdlib-only Python scripts.finance./

0 Comments

Login required
Log in to post a comment or update on this repo.

No comments yet — be the first to share an update.