# Geo-Parity (remote · geo-parity.com) Cost of living, fair global pay, healthcare costs, nomad visas and tax residency. Free, no key. - Trust score: 87/100 (high trust) - Change this week: +5 - Registry status: active - Liveness: live - Owner verified: no - Last scored: 2026-09-21 ## Components - remote · `geo-parity.com`: 87/100 (this document), [markdown](https://verifymcp.io/servers/com-geo-parity-geo-parity/api-mcp.md), [page](https://verifymcp.io/servers/com-geo-parity-geo-parity/api-mcp) ## Channel facts - Endpoint: `https://geo-parity.com/api/mcp` - Transports: `streamable-http` - Auth: `none` - Version: `0.2.0` ## Trust breakdown How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog) Scored 2026-09-21. - **Endpoint Security**: 83/100 - The endpoint's TLS certificate is valid, in date, and uses a strong key. - No authorisation is required to call this server. Every tool declares its destructiveHint and none is destructive, so open access doesn't expose one. - HTTPS is enforced; there's no plaintext access path. - The HSTS (Strict-Transport-Security) header is present. - DNSSEC is configured correctly; the domain's records validate against the full chain to the root. - **Transport & Reachability**: 100/100 - Verified streamable-http transport via a live MCP handshake. - **Schema Quality & AI Usability**: 85/100 - 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name). - AI-judged instruction clarity (excellent). - Context-footprint check failed: tool/resource definitions use about 18248 tokens (~276/item across 66 items; 62 tools + 4 resources), over budget; trim descriptions and params. - Tools include usage examples. - **Stability & Change Management**: 87/100 - Stability observed for 26 of 30 days with no destabilising changes; credit accrues until the full window elapses. - **Tool Coverage**: 98/100 - 100% of tools have a non-trivial description (not blank, and not just the tool's name). - 92% of tool parameters carry a description. - Structured output schemas are declared (98% of tools); any adoption earns full credit. - **Tool Safety**: 75/100 - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured. - 0 of 2 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "compare_occupation_pay" implies "pay" and declares readOnlyHint instead, contradicting what its own name says it does. - An AI judge read all 64 captured unit(s) of tool text and found none that tries to manipulate the model reading it. - **Capabilities**: 100/100 - Implements a current MCP spec version (2026-07-28). ## Install ### How do I install the Geo-Parity MCP server? Geo-Parity is a hosted endpoint at https://geo-parity.com/api/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation. ### Claude ```bash claude mcp add --transport http com-geo-parity-geo-parity 'https://geo-parity.com/api/mcp' ``` ### Cursor ```json { "mcpServers": { "com-geo-parity-geo-parity": { "url": "https://geo-parity.com/api/mcp" } } } ``` ### VS Code ```json { "servers": { "com-geo-parity-geo-parity": { "type": "http", "url": "https://geo-parity.com/api/mcp" } } } ``` ### Codex ```toml [mcp_servers.com-geo-parity-geo-parity] url = "https://geo-parity.com/api/mcp" ``` ### opencode ```json { "$schema": "https://opencode.ai/config.json", "mcp": { "com-geo-parity-geo-parity": { "type": "remote", "url": "https://geo-parity.com/api/mcp", "enabled": true } } } ``` ### OpenClaw ```bash openclaw mcp add com-geo-parity-geo-parity --url 'https://geo-parity.com/api/mcp' --transport streamable-http ``` ### Hermes ```yaml mcp_servers: com-geo-parity-geo-parity: url: "https://geo-parity.com/api/mcp" ``` ### Netclaw ```json { "McpServers": { "com-geo-parity-geo-parity": { "Transport": "http", "Url": "https://geo-parity.com/api/mcp" } } } ``` ### Vellum ```bash assistant mcp add com-geo-parity-geo-parity -t streamable-http -u 'https://geo-parity.com/api/mcp' ``` ### Other ```json { "mcpServers": { "com-geo-parity-geo-parity": { "type": "http", "url": "https://geo-parity.com/api/mcp" } } } ``` The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs. ## Changelog Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims. ### 2026-09-21 (score 87, +1) - [security] Tool “list_datasets” rewrote its description, which is the text the model reads - [security] Tool “passport_power” rewrote its description, which is the text the model reads ### 2026-09-20 (score 86, 0) - [functional] New tool “find_country_fit” ### 2026-09-19 (score 86, +2) - [functional improvement] Schema quality: fail → pass - [functional] New tool “compare_food_cost” ### 2026-09-18 (score 84, +1) - [functional] New tool “banking_access” ### 2026-09-15 (score 83, +1) No change was recorded against any check on this day. Stability & Change Management went from 63 to 67. That category is still filling its 30-day observation window: 19 days of observed history at the previous scan, 20 at this one. The score rises as the window fills, whether or not the server changes. ### 2026-09-14 (score 82, 0) - [security] Tool “country_dossier” rewrote its description, which is the text the model reads - [functional] New tool “driving_rules” - [functional] New tool “employment_rules” ### 2026-09-13 (score 82, +1) No change was recorded against any check on this day. Stability & Change Management went from 57 to 60. That category is still filling its 30-day observation window: 17 days of observed history at the previous scan, 18 at this one. The score rises as the window fills, whether or not the server changes. ### 2026-09-11 (score 81, +3) - [functional] The server now declares the “prompts” capability - [functional] The server now declares the “resources” capability - [functional] First check of Schema quality: 100 - [functional] New prompt “fair_local_salary” - [functional] New prompt “price_level_provenance” - [functional] New prompt “relocation_brief” - [functional] New prompt “remote_work_setup” - [functional] New prompt “retire_on_budget” - [functional] New resource “countries” - [functional] New resource “datasets” - [functional] New resource “exchange-regimes” - [functional] New resource “methodology-price-levels” ## MCP tools (62) ### `compare_total_cost` (~287 tokens) Total Cost of Living vs Your Income What a month actually costs in each of 41 countries on a stated income, and what is left over. Sums four lines that are kept from overlapping: rent (measured Eurostat city rents, HUD Fair Market Rents, or the price basket where neither covers a country), the everyday basket EXCLUDING rent (the basket normally contains rent — roughly half of it — so adding a separate housing figure to the full basket would overstate a total by ~60%), out-of-pocket healthcare, and income tax plus employee social contributions on that income. VAT is already inside the basket prices (purchasing-power data derives from surveyed purchaser prices) and is therefore never added. Tax assumes you become tax-resident and are taxed locally, single filer, no deductions; special regimes such as Portugal's IFICI are reported separately where they would be cheaper. Estimates for comparison — not a quote and not tax advice. Input parameters: - `country` (string): ISO 3166-1 alpha-2; echoes that country's row and rank as `focus`. - `income` (number, required): Monthly income in USD. Required — the tool is a subtraction. - `maxPctOfIncome` (number): Keep only countries whose total is at or below this share of income (%). - `sort` (string): Ranking. 'remaining' = most left over first; 'share' = smallest share of your income. ### `calculate_ppp` (~352 tokens) Calculate Purchasing Power Parity Given a US-denominated annual income and a target country code, returns the effective local purchasing power (income / pppIndex), the lifestyle equilibrium cost (income * pppIndex), and the purchasing power ratio (1 / pppIndex). Figures use real PPP data (World Bank, IMF, OECD, Penn World Table) relative to a US baseline of 1.0. Input parameters: - `city` (string): Optional city slug (e.g. 'london', 'san-francisco') for a sub-national Capital-to-Rural adjustment: multiplies the national price level by a per-city factor (curated overrides / US-metro BEA RPP / po… - `income` (number, required): Nominal annual income, denominated against the US baseline. - `source` (string): PPP dataset to draw figures from. 'consensus' = per-country median across the GDP-based providers (recommended default); 'worldbank' = World Bank GDP price level; 'imf' = IMF WEO implied PPP; 'oecd'… - `target` (string, required): ISO 3166-1 alpha-2 target country code (case-insensitive). Available codes depend on the chosen source. Output parameters: - `metrics` (object) - `ok` (boolean) - `source` (object) - `target` (object) ### `localize_salary` (~190 tokens) Localize a salary for fair pay across borders Convert a base salary from a home country into a cost-of-living-fair local equivalent: S_local = S_base × (I_target / I_home), where I is a US-relative price level. Omit `target` to receive a localized salary matrix across every country in the source — useful for pricing one offer across many markets. Returns USD purchasing-power equivalents, not currency-converted amounts. Input parameters: - `base` (number, required): Base annual salary in the home country (USD-denominated). - `home` (string, required): ISO 3166-1 alpha-2 home country code the base salary is expressed in (case-insensitive). - `source` (string): PPP dataset to draw figures from. Defaults to 'consensus' if omitted. - `target` (string): Optional ISO 3166-1 alpha-2 target country code. Omit to localize across all countries in the source. Output parameters: - `home` (object) - `input` (object) - `ok` (boolean) - `results` (array) - `source` (object) ### `calculate_wage_bands` (~240 tokens) Benchmark local salary bands (min, median, top 10%) Models a country's wage distribution as log-normal, calibrated from World Bank GNI per capita (mean) and the Gini index (spread: sigma = sqrt(2)*normInv((Gini+1)/2)), then returns the survival floor (essentials), statutory minimum (ILOSTAT where available, else curated; null where none exists), median (p50), comfortable (p75), top 10% (p90) and top 1% (p99) bands — each monthly and annual, USD and local currency. Optional `salary` locates a pay figure on the curve (percentile + out-earns %); `percentile` returns that exact threshold wage. Model estimates for benchmarking, not payroll data. Input parameters: - `country` (string, required): ISO 3166-1 alpha-2 country code (case-insensitive). - `percentile` (number): Return the exact wage threshold at this percentile (e.g. 90 for the top-10% floor). - `salary` (number): Your salary to locate on the distribution. Values over 20000 are read as annual (÷12), else monthly. Output parameters: - `atPercentile` (object) - `bands` (object) - `country` (object) - `model` (object) - `ok` (boolean) - `standing` (object) ### `compare_occupation_pay` (~255 tokens) Measured pay by occupation (ILO), and the gender gap in each MEASURED gross monthly earnings of employees by ISCO-08 major occupation group, from ILOSTAT — median and mean, in USD, PPP dollars and local currency, for 35 countries (a median is published for most of them). This is the site's only measured wage series; calculate_wage_bands is MODELLED, and the two are not interchangeable. Each row also carries the female-vs-male pay gap for that occupation (positive means women earn less; negative values are real and are kept) and where the occupation's median sits on the country's modelled national wage ladder. Employees only — excludes the self-employed and informal workers, which is why these diverge most from GNI-derived estimates in countries with large informal sectors. Before tax. The ten ISCO-08 major groups are broad families of work, not job titles: one group covers a surgeon and a junior developer alike. Countries with no ILO data return NO_DATA rather than a modelled substitute. Input parameters: - `country` (string, required): ISO 3166-1 alpha-2 country code (case-insensitive). - `occupation` (string): Narrow the response to one ISCO-08 major group. Omit for the whole table. Output parameters: - `basis` (string): Always 'measured' on success. - `country` (object) - `measuredMedianMonthlyUsd` (number) - `modelRatio` (number): measured / modelled for the country. Above 1: the model understates local pay. - `modelledMedianMonthlyUsd` (number) - `ok` (boolean) - `rows` (array) - `year` (number): ILO reference year — one year across every occupation in the country. ### `build_pay_range` (~287 tokens) Build a defensible pay range for a role in a country Returns a floor, midpoint and ceiling salary for a broad occupation and seniority in a country, anchored on ILO's measured earnings where they exist and on the modelled national wage ladder where they do not — `basis` says which. Built for the employer side of pay transparency: since 7 June 2026 the EU Pay Transparency Directive has required a starting salary or pay range in the job advert. IT IS NOT LEGAL ADVICE AND DOES NOT CERTIFY COMPLIANCE — it is a market range from official statistics, and every response carries a `caveats` array saying so, which should be reproduced alongside any figure quoted from it. Seniority is applied as a percentile offset within the country's own distribution rather than a fixed multiplier, so the same level means the same position locally rather than the same ratio. The gender pay gap for the occupation is reported beside the range and never folded into it. ISCO-08 major groups are broad families of work, not job titles. Input parameters: - `country` (string, required): ISO 3166-1 alpha-2 country code (case-insensitive). - `occupation` (string): ISCO-08 major group. Defaults to 'total' (all occupations). An unrecognised value is rejected, never widened to the default. - `seniority` (string): Defaults to 'mid'. An unrecognised value is rejected. Output parameters: - `basis` (string): 'measured' or 'modelled' — never blurred. - `caveats` (array): Always present. Reproduce these with any quoted figure. - `ceiling` (object) - `country` (object) - `floor` (object) - `genderGapPct` (number) - `midpoint` (object) - `occupation` (object) - `occupationMedianMonthlyUsd` (number) - `ok` (boolean) - `seniority` (object) - `year` (number) ### `check_schengen` (~167 tokens) Check a Schengen 90/180-day visa itinerary Validate a set of trips against the Schengen Area rule: a non-EU visitor may be present at most 90 days within any rolling 180-day window. Entry and exit days both count as days of presence; overlapping trips are de-duplicated. Returns days used in the current window, days remaining, an overall compliance flag, and the exact first overstay date (if any). Input parameters: - `plannedEntry` (string): Optional planned entry date. When set, the response adds plan: { maxStay, latestExit, fullResetDate } — the longest compliant stay starting on that date. - `referenceDate` (string): Optional 'as of' date (ISO YYYY-MM-DD). Defaults to today (UTC). - `trips` (array, required): List of Schengen stays. Output parameters: - `compliant` (boolean) - `daysRemaining` (number) - `daysUsed` (number) - `firstOverstayDate` (string|null) - `maxDaysInWindow` (number) - `ok` (boolean) - `referenceDate` (string) - `windowStart` (string) ### `find_meeting_slot` (~259 tokens) Score team timezone friction + find the meeting window Given a set of cities/timezones and each participant's local working hours, find the longest UTC window where everyone is simultaneously within their working day (DST-accurate), AND score the team's overall timezone friction. Returns the overlap in UTC and in each participant's local time (or null with the best partial coverage), plus a `friction` block: a 0–100 score, grade (Effortless/Manageable/Strained/Broken), daily shared hours, coverage, and the `loneWolf` — the member whose removal would free the most team overlap. Useful for scheduling and for hiring/team-composition decisions. Input parameters: - `date` (string): Reference date (ISO YYYY-MM-DD) for DST. Defaults to today (UTC). - `end` (number): Default working-window end, minutes from local midnight (default 1080 = 18:00). - `hours` (string): Optional per-zone overrides aligned to `zones`, e.g. '540-1080;420-720'. - `start` (number): Default working-window start, minutes from local midnight (default 540 = 09:00). - `zones` (array, required): IANA timezone identifiers, e.g. 'Europe/London'. Output parameters: - `bestPartialCount` (number) - `friction` (object) - `ok` (boolean) - `overlap` (object|null) - `referenceDate` (string) ### `compare_contract` (~276 tokens) Compare contractor vs employee net take-home Compute gross-to-net parity between a self-employed contractor hourly rate and an employee (W2/EOR) hourly rate, accounting for self-employment/income tax and self-paid health insurance, and return the break-even contractor rate. Uses transparent effective-rate tax presets per country, all overridable. Estimates only — not tax advice. Input parameters: - `contractorRate` (number, required): Self-employed hourly rate (USD). - `contractorWeeks` (number): Contractor paid/billable weeks (default 48). - `country` (string): Country code for tax presets. - `ctRate` (number): Override contractor effective tax rate, percent (e.g. 32). - `empRate` (number): Override employee effective tax rate, percent (e.g. 24). - `employeeRate` (number, required): Employee/W2/EOR hourly rate (USD). - `employeeWeeks` (number): Employee paid weeks (default 52). - `expenses` (number): Annual business expenses (USD, default 0). - `health` (number): Annual self-paid health insurance (USD); defaults to the country preset. - `hours` (number): Hours per week (default 40). - `taxMode` (string): 'brackets' uses progressive 2025 schedules (US and GB only; falls back to flat elsewhere — response echoes the mode). Output parameters: - `breakevenRate` (number) - `contractor` (object) - `contractorProfitable` (boolean) - `employee` (object) - `netDifference` (number) - `ok` (boolean) ### `select_region` (~200 tokens) Select the lowest-latency cloud host region Given a user-population breakdown, find the cloud host region that minimizes population-weighted round-trip latency (argmin_j Σ_i P_i × L_ij), and compute the optimal multi-region split (each user region routed to its nearest host). Latency is estimated from great-circle distance via a fibre-propagation model, not measured. Input parameters: - `hosts` (string): Optional comma-separated host region ids to restrict candidates (e.g. 'us-east-1,eu-west-2'). - `provider` (string): Optional cloud provider filter, applied before the hosts whitelist. - `users` (string, required): Comma-separated region:percent pairs (percentages auto-normalize), e.g. 'US-East:40,UK:30,Germany:30'. Region ids: US-East, US-West, Canada, Brazil, UK, Germany, France, Spain, Nigeria, South-Africa,… Output parameters: - `best` (object) - `ok` (boolean) - `ranking` (array) - `split` (object) ### `compare_countries` (~104 tokens) Compare two countries' cost of living Side-by-side comparison of two countries: US-relative price levels across every dataset (World Bank, IMF, OECD, Penn World Table, consumption basket), the consensus ratio, percent difference, and $100k salary equivalence in both directions. Input parameters: - `a` (string, required): First ISO 3166-1 alpha-2 country code (case-insensitive). - `b` (string, required): Second ISO 3166-1 alpha-2 country code (must differ from a). Output parameters: - `ok` (boolean) - `pctDifference` (number) - `ratioAtoB` (number) - `salary100k` (object) ### `required_rate` (~196 tokens) Compute the required freelance hourly rate Given a desired annual net income, compute the freelance hourly and day rate required after taxes, self-paid health insurance, weeks off, and billable utilization. Uses the same per-country effective-rate presets as compare_contract; all overridable. Estimates only — not tax advice. Input parameters: - `country` (string) - `desiredNet` (number, required): Desired annual net take-home (USD). - `expenses` (number): Annual business expenses (USD, default 0). - `health` (number): Annual self-paid health insurance (USD); defaults to the country preset. - `hours` (number): Hours per week (default 40). - `taxRate` (number): Override effective tax rate, percent (e.g. 32). - `utilization` (number): Billable share of working hours, 0–1 (default 0.75). - `weeksOff` (number): Unpaid weeks off per year (default 4). Output parameters: - `billableHours` (number) - `dayRate` (number) - `hourlyRate` (number) - `ok` (boolean) - `requiredGross` (number) ### `check_residency` (~88 tokens) Check tax-residency day counts Count distinct days of presence per country per calendar year and evaluate the generic 183-day rule, the US Substantial Presence Test (weighted 3-year formula), and the UK 183-day tax-year test. Early-warning day counter — not tax advice. Input parameters: - `referenceDate` (string): Optional 'as of' date. Defaults to today (UTC). - `stays` (array, required) Output parameters: - `countries` (array) - `ok` (boolean) ### `check_travel_residency` (~234 tokens) Unified Schengen + tax-residency + tax-cost tracker The comprehensive engine: from one travel itinerary it returns (1) Schengen 90/180 tourist-visa status (days used/remaining, first overstay, longest compliant stay), (2) per-country tax-residency risk on each country's OWN threshold and basis — the 183-day rule, Cyprus's conditional 60-day rule (flagged only when you're not 183+ resident elsewhere), the US Substantial Presence Test and the UK tax-year test — with safe days left, leave-by dates and a safe/warning/triggered rating, and (3) with an optional income, the tax cost of triggering residency at the standard rate vs an expat regime. Supersedes check_schengen / check_residency / check_tax_residency_risk (all still available). Early-warning day math — not tax advice. Input parameters: - `income` (number): Optional gross annual income (USD) — enables the tax-cost projection for at-risk countries. - `referenceDate` (string): Optional 'as of' date. Defaults to today (UTC). - `stays` (array, required) Output parameters: - `countries` (array) - `ok` (boolean) - `referenceDate` (string) - `schengen` (object|null): 90/180 status when any Schengen stay exists: daysUsed, daysRemaining, compliant, firstOverstayDate, maxStayFromRef. - `summary` (object) ### `find_fire_locations` (~267 tokens) Find where a portfolio can fund early retirement Geo-arbitrage FIRE analysis: given a portfolio and safe withdrawal rate, returns every country where the passive income sustains a lean ($32k US-equivalent/yr), comfortable ($55k), upper ($90k), or luxury ($150k) lifestyle at local purchasing-power prices — plus FIRE numbers per tier, nomad-visa income-floor eligibility, local-currency income, an SWR sensitivity band, (with savings/coast params) years until each country unlocks, and — where covered — a `statePension` overlay with the OECD net pension replacement rate (safety net from pension age; not subtracted from the FIRE numbers). Estimates only — not financial advice. Input parameters: - `coast` (string): Set to 1 for Coast-FIRE mode (project growth with zero further savings). - `portfolio` (number, required): Total investable portfolio (USD). - `return` (number): Expected real annual return, fraction (0.05) or percent (5). Default 0.05. - `savings` (number): Monthly savings (USD) — enables the years-to-FIRE planner. - `source` (string) - `swr` (number): Safe withdrawal rate as fraction (0.04) or percent (4). Default 0.04. Output parameters: - `annualIncome` (number) - `monthlyIncome` (number) - `ok` (boolean) - `rows` (array) - `sensitivity` (array) ### `estimate_micro_costs` (~254 tokens) Estimate everyday prices in two countries (Pint & Property Index) Estimates local prices for a 32-item everyday basket (rent, pint of beer, groceries, transport, gym, …) in two countries, derived as usPrice × pppIndex^elasticity — non-tradables like rent swing harder across borders than world-priced goods like petrol (Balassa-Samuelson-style model). Returns per-item prices in USD and local currency, category subtotals, and a single-person monthly budget rollup. Model estimates, not price quotes. Input parameters: - `base` (string, required): ISO 3166-1 alpha-2 country code to compare FROM (case-insensitive). - `baseCity` (string): Optional city slug for the base side (sub-national cost adjustment, e.g. 'london'). Enables intra-country city comparisons like London vs Hull. - `source` (string): Optional dataset override. Defaults to the consumption basket (worldbank-cons) when it covers both countries, else consensus; the response echoes the source used. - `target` (string, required): ISO 3166-1 alpha-2 country code to compare TO (must differ from base). - `targetCity` (string): Optional city slug for the target side (e.g. 'hull'). Output parameters: - `categoryTotals` (array) - `items` (array) - `monthlyBudget` (object) - `ok` (boolean) - `source` (object) ### `check_tax_residency_risk` (~214 tokens) Check tax-residency risk for remote work abroad Remote Worker Tax Trap radar: evaluates how close a stay is to triggering tax residency (183-day rule with per-country variations across 43 tax jurisdictions) and estimates the tax owed on the FULL annual income if the threshold is crossed — comparing the standard resident effective rate against the country's special expat/nomad regime (Spain's Beckham Law 24%, Portugal's IFICI 20%, Italy's impatriati, territorial/remittance systems, …). Optionally computes leave-by/residency dates from an arrival date and warns where the Schengen 90/180 tourist limit bites first. Early-warning estimates — not tax advice. Input parameters: - `arrival` (string): Optional arrival date — adds leaveBy and residencyDate to the response. - `country` (string, required): ISO 3166-1 alpha-2 country code (case-insensitive). - `days` (number, required): Days of presence in the relevant counting period. - `income` (number, required): Gross annual income (USD) the residency cliff would expose. Output parameters: - `dates` (object) - `daysRemaining` (number) - `ok` (boolean) - `pctOfThreshold` (number) - `projection` (object) - `risk` (string) - `schengenWarning` (string) - `thresholdDays` (number) ### `rank_livability` (~227 tokens) Rank countries by quality of life vs cost Quality-of-Life vs Cost matrix: ranks 41 countries by a user-weighted composite of livability metrics (median broadband Mbps, safety, healthcare, air quality, English usability — curated country-level data replacing the discontinued Teleport API) against the monthly cost of single-person essentials at local purchasing-power prices. Returns per-country composite and value scores, lifestyle label for your budget (comfortable → like royalty), nomad-visa eligibility, and days-to-tax-residency. Curated estimates — verify locally. Input parameters: - `budget` (number): Monthly budget (USD) — drives lifestyle labels and visa eligibility. - `maxBudget` (number): Filter: keep countries whose monthly essentials cost ≤ this (USD). - `minInternet` (number): Filter: minimum median broadband, Mbps (e.g. 80). - `minSafety` (number): Filter: minimum safety score, 0–100. - `weights` (string): Five comma-separated 0–10 weights: internet,safety,healthcare,air,english (default 5,5,5,5,5). Output parameters: - `excluded` (number) - `ok` (boolean) - `rows` (array) ### `compare_healthcare` (~403 tokens) Compare Healthcare Costs & Exposure Healthcare cost and financial exposure across 41 countries: out-of-pocket spend per person per year (US$), out-of-pocket as a share of all health spending, total health spend per capita (PPP international $), government funding share, physicians per 1,000 and dentists per 10,000. Figures are MEASURED from World Bank and WHO Global Health Observatory series, never derived from a price index: scaling healthcare by a country's general price level overstates costs abroad by 1.2x-5.4x, because health prices do not track the price level (the US is an outlier and several countries deliberately price healthcare below their own economies). Share and burden are separate facts and both are returned - the US pays only ~11% out of pocket, a low share, but the highest absolute amount. Dental is reported as ACCESS (dentist density), not price: no free citable source publishes procedure prices. National averages, not a quote for any individual's care. Not medical, insurance or financial advice. Input parameters: - `country` (string): Optional ISO 3166-1 alpha-2 country code (case-insensitive). Echoes that country's row and its rank under the active sort. - `maxOutOfPocketPct` (number): Optional filter: keep only countries where households pay at most this percentage of health spending directly. - `minPhysiciansPer1000` (number): Optional filter: keep only countries with at least this many physicians per 1,000 people. - `sort` (string): Metric to rank by. 'outOfPocket' = out-of-pocket spend per person per year (default, the figure a person actually pays); 'spend' = total health spend per capita; 'exposure' = out-of-pocket as a share… Output parameters: - `attribution` (object): Sources, a quotable citation and a permalink. - `excluded` (number): Countries removed by filters or missing the sorted metric. - `focus` (object): The requested country's row, when `country` was given. - `ok` (boolean) - `rows` (array): Countries under the active sort, best first. - `updated` (string): ISO date the underlying data was last refreshed. ### `compare_housing` (~445 tokens) Compare Housing Costs & Affordability Housing cost and affordability across countries, built for remote workers. Returns monthly rent plus TWO ratios that together answer the real question: rent as a share of the LOCAL median wage (are locals priced out?) and as a share of YOUR remote income (can you afford it?). The gap between them - `arbitrageGap`, in percentage points - is the single number saying whether a place is cheap FOR YOU; pass `remoteIncome` to get it. Also returns price-to-income and price-to-rent as a percentage of each country's OWN long-run average (OECD), and mortgage rates (ECB, euro area only; absent elsewhere rather than estimated). IMPORTANT: those affordability percentages compare a country with its own history, NOT with other countries - 127 for Spain means 27% less affordable than Spain's norm, not 27% dearer than Germany. Every rent carries `rentBasis`: 'measured' where a real Eurostat city figure exists (Spain, Portugal, Switzerland only) or 'modelled' from the site's price basket, which understates housing by up to ~40% where housing has decoupled from the general price level. National averages; not financial advice. Input parameters: - `country` (string): Optional ISO 3166-1 alpha-2 code; echoes that country's row and rank. - `flatM2` (number): Flat size in m² used to convert measured per-m² rents. Default 70; echoed back so the assumption is visible. - `maxRentPctOfIncome` (number): Optional filter: keep only countries where rent is at most this percentage of your remote income. Requires remoteIncome. - `remoteIncome` (number): Your monthly income in USD, earned remotely. Enables rentPctOfYourIncome and arbitrageGap; omit for the local-only view. - `sort` (string): 'affordability' = rent as a share of your remote income, lowest first (falls back to raw rent when no income given); 'rent' = monthly rent; 'localBurden' = rent as a share of the local median wage; '… Output parameters: - `attribution` (object) - `excluded` (number) - `focus` (object) - `ok` (boolean) - `rows` (array) - `updated` (string) ### `find_nomad_visas` (~222 tokens) Match income to digital nomad visas Digital-nomad visa Income Matcher: given a monthly remote income (any currency) and household size, computes each program's real income threshold (base plus its published per-dependent uplift) and returns the programs you qualify for plus the near misses just out of reach — with duration, renewability, tax notes and official source URLs. Curated summaries — verify with the official source before applying. Input parameters: - `children` (number): Number of dependent children (each adds the program's child uplift). - `currency` (string): ISO currency code for `income` (e.g. EUR, GBP, INR). Defaults to USD. - `income` (number): Your gross monthly income (in `currency`, default USD). - `minDuration` (number): Minimum program duration in months. - `partners` (number): 1 if bringing a partner (raises each program's threshold by its partner uplift). - `renewable` (string): Set to 1 to require renewable programs. - `residency` (string): Set to 1 to require a path to permanent residency. Output parameters: - `household` (object) - `incomeUsd` (number): Your income converted to USD, used for matching. - `matches` (array): Programs you qualify for; each carries requiredUsd (household threshold) and shortfallUsd (≤0). - `nearMisses` (array): Programs within ~$1,000/mo of qualifying, closest first. - `ok` (boolean) - `total` (number) - `updated` (string) ### `passport_power` (~123 tokens) Passport visa-free access score and global rank Passport Power: for a given country's passport, returns the visa-free-access score (visa-free + visa-on-arrival + eTA destinations), global rank and percentile among all ~199 passports, and the full destination breakdown by access type with names resolved. Omit country for the ranked league table. Reference data from the open imorte/passport-index-data (public domain) — confirm entry rules with the destination authority before travel. Input parameters: - `country` (string): Passport country alpha-2 code (e.g. US, GB, SG). Omit for the full league table. Output parameters: - `ok` (boolean) - `passport` (object) - `percentile` (number) - `resolved` (object) - `totalPassports` (number) ### `calculate_runway` (~266 tokens) Runway multiplier from relocating (bootstrap buffer) For founders/freelancers stretching savings: computes how many extra months of runway relocating to a cheaper hub buys. Burn is split into fixed (location-independent — subscriptions, debt, insurance) and local; only the local slice scales by the target's cost-of-living ratio (city-level where a curated hub is given, else national, incl. Factbook-only countries). Returns the home baseline plus each hub's monthly burn, runway months, extra months, run-out date and the ×multiplier. Estimates from purchasing-power data, not financial advice. Input parameters: - `burn` (number, required): Total current monthly burn at home (in `currency`). - `currency` (string): ISO code for the money inputs (e.g. GBP). Default USD. - `fixed` (number): Location-independent slice of burn (0…burn) that won't fall when you move. - `home` (string): Home country alpha-2 code. Default US. - `homeCity` (string): Optional curated home-city slug. - `savings` (number, required): Total savings (in `currency`). - `targets` (string, required): 1–3 hubs, comma-separated: COUNTRY or COUNTRY:citySlug (e.g. TH:chiang-mai,AR:buenos-aires,GE). Output parameters: - `fixedFloorMonths` (number|null): Runway cap even at zero local cost (savings ÷ fixed burn). - `home` (object) - `hubs` (array): Sorted by extra months; each has runwayMonths, extraMonths, multiplier, monthlyBurnUsd, runsOut, priceRatio and a coverage note when approximate. - `ok` (boolean) ### `compare_connectivity` (~227 tokens) Compare measured internet quality by country Measured connection quality per country from Cloudflare Radar: median download and upload (Mbps), median idle latency and jitter (ms), and a call-quality verdict derived from latency and jitter rather than bandwidth — a video call fails on latency and jitter long before it fails on throughput, which is why a download figure alone cannot answer “can I work from here”. Aggregated from real connections over a 28-day window, not crowdsourced or self-reported. Coverage is wider than the site's country pages (~200 locations), so this answers for places with no page here. A country Radar has too little data for is absent, never zero. Input parameters: - `limit` (number): Return at most this many countries. - `maxLatency` (number): Filter: maximum median idle latency, ms (e.g. 100). - `minDownload` (number): Filter: minimum median download, Mbps (e.g. 50). - `sort` (string): Metric to rank by. Bandwidth sorts high-to-low; latency and jitter sort low-to-high, because “better” runs in opposite directions. Output parameters: - `ok` (boolean) - `rows` (array) - `sort` (string) - `updated` (string) ### `compare_public_holidays` (~254 tokens) Shared working days and public holidays by country Two questions in one. Given `countries` (two or more), returns the working days those places genuinely SHARE in a year — weekdays on which nobody is on a public holiday — which is the number a distributed team actually plans against and the half of the question timezone-overlap tools never answer: a London/Tokyo team loses roughly three working weeks a year to non-overlapping holidays. Without `countries`, ranks countries by nationwide public holidays. Weekend holidays are excluded (they cost nobody a working day) and regional holidays are counted separately from national ones. Covers 204 countries, but NOT India, Thailand, Malaysia, the UAE or Saudi Arabia — those return NO_DATA rather than an empty calendar, because “no data” and “no holidays” are opposite claims. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more (e.g. "GB,JP"). Omit to get a ranking instead of an overlap. - `limit` (number): Return at most this many countries in a ranking. - `sort` (string): Ranking direction. Ignored when `countries` is given. - `year` (number): Calendar year. Defaults to the first year in the dataset. Output parameters: - `countries` (array) - `lostDates` (array) - `lostDays` (number) - `ok` (boolean) - `rows` (array) - `sharedWorkingDays` (number) - `weekdays` (number) - `year` (number) ### `plan_team_calendar` (~401 tokens) Team calendar: shared working days and hours across countries When a distributed team can actually all work. Joins the two halves no free tool puts together: the weekdays on which NOBODY is on a public holiday (the engine behind compare_public_holidays) and the daily working-hour overlap across the members' time zones (the engine behind find_meeting_slot), evaluated on 15 January and 15 July so daylight-saving asymmetry is visible, and multiplied into conservative shared working hours per year. Members are ISO alpha-2 country codes; each country is placed in the time zone of its representative city, and countries spanning several zones (US, CA, BR, MX, AU, ID) are flagged `multiZone` — append `@IANA/Zone` to override (e.g. `US@America/Chicago`). Covers the 41-country spine EXCEPT India, Thailand, Malaysia, the UAE and Saudi Arabia, which have no published holiday calendar and return NO_DATA naming them rather than a calendar with no holidays. Working-hour windows are a stated assumption (09:00-18:00 local unless given), not a measurement; annual leave is not modelled. Input parameters: - `end` (number): Default local working-window end, minutes from midnight (default 1080 = 18:00). - `hours` (string): Per-member `start-end` overrides, semicolon-separated and aligned with `members`; leave a slot empty to keep the default (e.g. "540-1080;;420-720"). - `members` (string, required): Comma-separated ISO alpha-2 codes, two or more, each optionally `CC@IANA/Zone` (e.g. "GB,JP,US@America/Chicago"). - `start` (number): Default local working-window start for every member, minutes from midnight (default 540 = 09:00). - `year` (number): Calendar year for the holiday calendar. Defaults to the first year in the dataset. Output parameters: - `attribution` (object) - `caveats` (array) - `days` (object) - `hours` (object) - `members` (array) - `ok` (boolean) - `sharedWorkingHoursPerYear` (number) - `year` (number) ### `estimate_employer_cost` (~348 tokens) Employer cost of a hire, by country What a hire actually costs the employer: the gross wage plus the employer's own social security contributions at the rate OECD Taxing Wages publishes (single earner, no children, 100% of the average wage), applied as a flat rate — the figure employer-of-record services quote for a fee. Pass `gross` (annual USD) for your own figure; without it the wage is the pay-range midpoint for the `occupation` and `seniority`, and `basis` says whether that wage is measured (ILOSTAT) or modelled. Employee-side rates and OECD's tax wedge are returned for context and never applied. Without `country`, ranks the spine by employer contribution rate. Coverage is the 25 spine countries OECD measures; the other sixteen return NO_DATA, never a 0% rate (New Zealand's 0% is real: it has no employer contributions). Every response carries caveats — contribution ceilings, excluded schemes — and it is not payroll or legal advice. Input parameters: - `country` (string): ISO alpha-2 code (e.g. "DE"). Omit for a ranking. - `gross` (number): Gross annual wage in USD. When omitted, the pay-range midpoint is used. - `limit` (number): Return at most this many countries in a ranking. - `occupation` (string): ISCO-08 major group id for the default wage (default `total`). Ignored when `gross` is given. - `seniority` (string): Seniority for the default wage (default `mid`). Ignored when `gross` is given. - `sort` (string): Ranking direction. Ignored when `country` is given. Output parameters: - `attribution` (object) - `caveats` (array) - `country` (object) - `employee` (object) - `employer` (object) - `oecd` (object) - `ok` (boolean) - `rows` (array) - `taxWedge` (number) - `wage` (object) ### `compare_working_time` (~322 tokens) Working hours and paid parental leave, by country Two questions in one. Given `countries` (two or more ISO alpha-2 codes), returns side by side: mean weekly hours actually worked per employed person (ILOSTAT, newest year from 2023), average annual hours actually worked (OECD, members only), and the paid leave the law gives mothers and fathers in calendar days with who pays (World Bank Women, Business and the Law 2026). Without `countries`, ranks the 41-country spine by one `metric` — weeklyHours, annualHours, maternityDays or paternityDays; `sort=fewest` inverts. A country missing a metric lacks the key, and is dropped from a ranking rather than scored zero. OECD's annual-hours series is NOT comparable across countries, by OECD's own statement — the wording travels with every response that carries the figure. Weekly hours average full- and part-time workers together. Leave is what the law provides, not take-up; a legal score below 1 marks a partial provision. Statutory minimum ANNUAL leave is deliberately absent: no citable structured source exists. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more (e.g. "DE,US,TH"). Omit to get a ranking instead. - `limit` (number): Return at most this many countries in a ranking. - `metric` (string): Ranking metric. Ignored when `countries` is given. - `sort` (string): Ranking direction. Ignored when `countries` is given. Output parameters: - `attribution` (object) - `caveats` (array) - `metric` (string) - `ok` (boolean) - `rows` (array) ### `compare_inflation` (~282 tokens) Inflation, interest and real rates by country What money earns or loses in real terms. Given `countries` (two or more ISO alpha-2 codes), returns side by side: consumer-price inflation (World Bank, latest annual, plus the IMF's monthly year-on-year figure where available), the central-bank policy rate (BIS, newest month; the ECB's single rate for euro-area members, flagged `viaEuroArea`), deposit and lending rates (World Bank, where published), and the real rates the engine computes from them by the exact Fisher relation — only when the nominal rate and the inflation figure are within a year of each other. Without `countries`, ranks the 41-country spine by one `metric`. A country missing a metric lacks the key and is dropped from a ranking, never scored 0. Real rates are ex post (last year's inflation), not expectations; a projection at the latest rate is flat, not a forecast. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more (e.g. "GB,DE,TR"). Omit to get a ranking instead. - `limit` (number): Return at most this many countries in a ranking. - `metric` (string): Ranking metric. Ignored when `countries` is given. - `sort` (string): Ranking direction. Ignored when `countries` is given. Output parameters: - `attribution` (object) - `caveats` (array) - `metric` (string) - `ok` (boolean) - `rows` (array) ### `remittance_cost` (~302 tokens) Cost of sending money home, by corridor What it costs to send money from one country to another, from the World Bank's Remittance Prices Worldwide survey (newest quarter, CC BY 4.0): the mean total cost — fee plus exchange-rate margin — of sending USD 200 and USD 500 across the providers surveyed on that corridor, the cheapest provider found, and how many were surveyed. With `amount` (USD), a quote at the nearer of the two surveyed bands (below USD 350 uses the 200 figures), and the response says which. Give `from` AND `to` for one corridor; only one of them for a ranking of surveyed corridors out of a sender or into a receiver, cheapest first. The survey covers ~350 high-volume corridors (~46 sending countries, ~96 receiving); a pair it does not survey returns NO_DATA, never a figure borrowed from a neighbour. Your own quote will differ by provider, payment method and day. Input parameters: - `amount` (number): Amount to send, USD. Optional; without it the corridor's survey figures are returned unquoted. - `from` (string): Sending country, ISO alpha-2 (e.g. "GB"). - `limit` (number): Return at most this many corridors in a ranking. - `sort` (string): Ranking direction when only one of `from`/`to` is given. - `to` (string): Receiving country, ISO alpha-2 (e.g. "IN"). Output parameters: - `attribution` (object) - `avgCost200Pct` (number) - `avgCost500Pct` (number) - `caveats` (array) - `cheapestCost200Pct` (number) - `corridor` (object) - `ok` (boolean) - `period` (string) - `providers` (number) - `quote` (object) - `rows` (array) ### `pension_abroad` (~313 tokens) What a pension is worth abroad, and how fast it erodes A fixed pension paid from one country and spent in another, after the exchange rate, the destination's price level and the destination's inflation. Converts `amount` in `currency` at the site's FX snapshot, re-prices it at the destination through the PPP engine (the same price-level comparison as calculate_ppp), and projects it flat at the destination's latest annual inflation for `years` (1-40, default 10), returning what it is worth there today and after the period, in USD and local currency, with the share of purchasing power lost. Without `amount`, the origin's typical state pension is used — OECD's net replacement rate for an average earner on the modelled median wage — and `basis` says so; 33 origins have a rate, the rest need `amount`. Holds the pension and the exchange rate fixed and says so; many state pensions are uprated, and some are frozen for expatriates in particular countries. Not financial advice. Input parameters: - `amount` (number): Monthly pension in `currency`. Omit for the origin's typical state pension. - `currency` (string): ISO 4217 code of `amount` (default USD). - `from` (string, required): Country the pension is paid from, ISO alpha-2 (e.g. "GB"). - `to` (string, required): Country it will be spent in, ISO alpha-2 (e.g. "PT"). - `years` (number): Years to project (whole number, default 10). Output parameters: - `attribution` (object) - `caveats` (array) - `erosionPct` (number) - `from` (object) - `inflationPct` (number) - `ok` (boolean) - `pension` (object) - `to` (object) - `worthAfterYearsLocal` (number) - `worthAfterYearsUsd` (number) - `worthTodayLocal` (number) - `worthTodayUsd` (number) - `years` (number) ### `compare_banking_access` (~245 tokens) Banking access by country (Findex) Whether adults there can bank: the share of adults (15+) with an account at a bank, another financial institution or a mobile-money provider, from the World Bank's Global Findex — a triennial household survey, so the survey year travels with each figure — plus personal remittances received as a share of GDP. Given `countries` (two or more), side by side; without, the 41-country spine ranked by `metric` (`account` or `remittancesReceived`). Account ownership says whether adults CAN bank, not whether a newcomer can open an account — that depends on residency and documentation rules this tool does not cover. A missing figure is absent and dropped from rankings, never 0. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more (e.g. "GB,NG,IN"). Omit to get a ranking instead. - `limit` (number): Return at most this many countries in a ranking. - `metric` (string): Ranking metric. Ignored when `countries` is given. - `sort` (string): Ranking direction. Ignored when `countries` is given. Output parameters: - `attribution` (object) - `caveats` (array) - `metric` (string) - `ok` (boolean) - `rows` (array) ### `banking_access` (~434 tokens) Banking access — opening an account, deposit protection and instant payments Whether a newcomer can open a personal bank account in a country (`yes`, `restricted` or `no`) with the documents usually asked for from a fixed vocabulary, the statutory deposit guarantee per depositor per bank in local currency AND converted to USD at the site's FX snapshot, the country's instant payment systems from the World Bank's Project FASTT tracker, the EU basic-account right under Directive 2014/92/EU where it applies, and a FATCA flag where banks commonly decline US citizens — a hand-curated register over all 186 spine countries, added 17 Sep 2026, with the scheme or regulator page and check date per row (`confidence: "secondary"` where the official page could not be fetched, separately on `depositGuarantee` for the figure). A country with no deposit-insurance scheme has NO `depositGuarantee` — never a zero — and a country absent from FASTT has no `instantPayments`, which is not the same as having no instant payments. `country` for one entry, `countries` for a side-by-side; without either, ranks by `metric` (`depositGuaranteeUsd` default, or `instantPaymentsSince`, whose `sort=lowest` lists the longest-running first), optionally narrowed by `access`, with countries lacking the metric in `excluded`. `internationalOptions` describes the kinds of cross-border account and what protects the money — categories only, never a named provider. For information only — not financial advice; banks set their own onboarding policies and every response says so first. Input parameters: - `access` (string): Narrow a ranking to one non-resident access rule. - `countries` (string): Comma-separated ISO alpha-2 codes, at least two, e.g. GB,TH,NG. - `country` (string): ISO alpha-2 code for a single country, e.g. PT. - `limit` (number): Maximum rows in a ranking. - `metric` (string): Ranking metric when no country is given. - `sort` (string): Defaults to highest, or lowest for instantPaymentsSince (longest-running first). Output parameters: - `attribution` (object) - `caveats` (array) - `excluded` (array) - `internationalOptions` (array) - `ok` (boolean) - `rows` (array) ### `compare_tax_rates` (~310 tokens) Statutory corporate and top income-tax rates by country Statutory tax schedules from the OECD, compared or ranked. Given `countries` (two or more ISO alpha-2 codes), returns side by side the corporate income-tax rate (combined central plus average sub-national, central only, and a small-business rate where one exists; OECD Corporate Tax Statistics, all 41 spine countries) and, for OECD members, the top personal income-tax rate with the earnings threshold at which it starts as a multiple of the average wage, plus the marginal rate with employee social contributions at that point (OECD Tax Database, 26 of the spine). Without `countries`, ranks the spine by one `metric`. A country OECD's PIT flow does not cover has no `personal` block and is dropped from a PIT ranking, never scored 0. Rates are statutory, not effective: incentives, surtaxes and special regimes are outside the data, and this does not replace the blended take-home profiles behind `compare_contract`. Statutory figures for information only — not legal or tax advice; every response says so first. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more (e.g. "DE,IE,TH"). Omit to get a ranking instead. - `limit` (number): Return at most this many countries in a ranking. - `metric` (string): Ranking metric. Ignored when `countries` is given. - `sort` (string): Ranking direction. Ignored when `countries` is given. Output parameters: - `attribution` (object) - `caveats` (array) - `metric` (string) - `ok` (boolean) - `rows` (array) ### `legal_profile` (~364 tokens) Women, Business and the Law: what the statute says, by country The World Bank's Women, Business and the Law reading of each country's statute (2026 edition, 41 spine countries). Give `country` for one country: the Legal Frameworks index (0-100), eight topic scores (mobility, workplace, pay, marriage, parenthood, entrepreneurship, assets, pension), the coded answers grouped by topic — can a woman choose where to live, travel abroad, get a job the same way as a man, is equal pay for work of equal value mandated, can she obtain a divorce and head a household, own and inherit equally, open a business and access credit without discrimination — and the full, partial and mandatory retirement ages for men and women. Give `countries` (two or more) for the index, topics and ages side by side, or neither for the spine ranked by `metric` (`index`, a topic, `retirementAgeMen`, `retirementAgeWomen`). A partial answer keeps WBL's fraction as `{partial}` rather than rounding to yes or no; a true says the provision exists in law, not that it is enforced. Statutory figures for information only — not legal or tax advice; every response says so first. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more, for a side-by-side comparison. Ignored when `country` is given. - `country` (string): One ISO alpha-2 code for the full profile (e.g. "SA"). - `limit` (number): Return at most this many countries in a ranking. - `metric` (string): Ranking metric when neither `country` nor `countries` is given. - `sort` (string): Ranking direction. Ignored when `countries` is given. Output parameters: - `attribution` (object) - `caveats` (array) - `edition` (number) - `index` (number) - `law` (object) - `ok` (boolean) - `retirement` (object) - `rows` (array) - `topics` (object) ### `compare_vat` (~293 tokens) VAT and GST rates by country, with the authority page each was read from Standard and reduced VAT/GST rates for the 41-country spine, hand-curated from each national tax authority's published page and served with that page's URL and the date it was checked, so a figure can be verified rather than trusted. Given `countries` (two or more ISO alpha-2 codes), returns each country's tax kind and name, standard rate, reduced rates where the authority lists them, registration threshold where known, and its source; an entry read from a secondary source because the authority's page could not be fetched is marked `confidence: "secondary"`. Without `countries`, ranks standard rates lowest first. Countries with no single national rate — the US (state sales taxes), Brazil (state and federal taxes in transition to a dual VAT) and Malaysia (sales and service tax) — are described by `kind`, excluded from the ranking and named in `excluded`, never given 0%. Reviewed at least yearly; not a live feed. Statutory figures for information only — not legal or tax advice; every response says so first. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more (e.g. "GB,US,DE"). Omit to get a ranking instead. - `limit` (number): Return at most this many countries in a ranking. - `sort` (string): Ranking direction. Ignored when `countries` is given. Output parameters: - `attribution` (object) - `caveats` (array) - `excluded` (array) - `ok` (boolean) - `rows` (array) ### `travel_advice` (~339 tokens) UK and US government travel advice, by country What two governments say about going. Give `country` (ISO alpha-2) for one country: the UK FCDO's level (none, all-but-essential to parts, all-but-essential, all travel to parts, all travel) with its latest change and date, the FCDO's passport-validity and visa sentences for British citizens and the health risks it lists, and the US State Department's advisory level (1 exercise normal precautions to 4 do not travel) with its headline. Give `countries` (two or more) for the levels side by side, or neither for the 41-country spine ranked least-restricted first by FCDO level then US level (`sort=highest` inverts). The two governments grade differently and often disagree; neither is rescored. The FCDO publishes nothing for the UK and the State Department nothing for the US, so those blocks are absent with a caveat, never a made-up level, and a country with neither is excluded from the ranking. Official sources on the dates shown, for information only — conditions change; check the source before travelling. Every response says so first. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more, for a side-by-side. Ignored when `country` is given. - `country` (string): One ISO alpha-2 code for the full advice (e.g. "TH"). - `limit` (number): Return at most this many countries in a ranking. - `sort` (string): Ranking direction when neither `country` nor `countries` is given: lowest = least restricted first. Output parameters: - `attribution` (object) - `caveats` (array) - `excluded` (array) - `fcdo` (object) - `ok` (boolean) - `rows` (array) - `us` (object) ### `health_prep` (~264 tokens) Vaccines, yellow-fever certificate and malaria, by country What to sort out with a travel clinic, as facts from TravelHealthPro (NaTHNaC, the UK's national travel-health centre) with the source page linked per row: the yellow-fever certificate position (required for all arrivals, required when arriving from a risk country, or none), whether the country itself carries yellow-fever risk, the vaccine names recommended for most UK travellers and for some, and whether malaria guidance exists — plus the health-risk names the FCDO lists. Give `country` for one country or `countries` (two or more) for a side-by-side. Facts only, never NaTHNaC's prose; the reasoning is at the linked page. These are recommendations for UK travellers and depend on itinerary and medical history — a clinic decides for a person. The UK itself is NO_DATA (no FCDO page, so no link), never an empty list. Official sources on the dates shown, for information only — conditions change; check the source before travelling. Every response says so first. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more. Ignored when `country` is given. - `country` (string): One ISO alpha-2 code (e.g. "BR"). Output parameters: - `attribution` (object) - `caveats` (array) - `malaria` (boolean) - `ok` (boolean) - `rows` (array) - `vaccinesMost` (array) - `vaccinesSome` (array) - `yellowFeverCertificate` (string) ### `compare_safety_stats` (~313 tokens) Homicide rate, PM2.5 and disaster risk by country Measured danger once there. Given `countries` (two or more ISO alpha-2 codes), returns side by side: intentional homicides per 100,000 (UNODC via World Bank WDI) and mean annual PM2.5 exposure in µg/m³ (WDI), each with its observation year, and the INFORM disaster-risk score and band composed for context. Without `countries`, ranks the 41-country spine by one `metric` (`homicide` default, `pm25`), lowest first. Only observations at or after the newest WDI vintage are held; a country whose newest figure is older lacks the key and is dropped from the ranking, never scored 0 — the UK's newest homicide observation predates the floor, so it has no homicide figure here. Homicide counts intentional killings only, not petty crime or road deaths; PM2.5 is a national population-weighted mean. Official sources on the dates shown, for information only — conditions change; check the source before travelling. Every response says so first. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more (e.g. "DE,BR,ZA"). Omit to get a ranking instead. - `limit` (number): Return at most this many countries in a ranking. - `metric` (string): Ranking metric. Ignored when `countries` is given. - `sort` (string): Ranking direction. Ignored when `countries` is given. Output parameters: - `attribution` (object) - `caveats` (array) - `metric` (string) - `ok` (boolean) - `rows` (array) ### `compare_pension_systems` (~373 tokens) State pension systems: retirement ages, replacement rates, years in retirement, by country What the state pension system looks like there, from OECD Pensions at a Glance (members and G20 partners, 33 of the 41 spine countries). Give `country` for one: normal retirement age now and for a 22-year-old entering today, by sex; net and gross replacement rates for a full-career average earner at 50/100/200% of the average wage under mandatory schemes; expected years after leaving work; the share of pensioners below half of median income; the average tax rate on an average pension; the old-age to working-age ratio and public pension spending; plus World Bank life expectancy and the 65+ share for all 41 — and WBL's statutory retirement age beside them, which stands alone for the eight countries OECD does not cover (no replacement rate is ever borrowed). Give `countries` (two or more) for a side-by-side, or neither for the spine ranked by `metric`. Projection rows (2060+) are excluded. A country missing a metric lacks the key and is dropped from a ranking, never scored 0. Published rules and modelled figures on the dates shown, for information only — entitlements depend on a person's record and consulates decide; every response says so first. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more, for a side-by-side. Ignored when `country` is given. - `country` (string): One ISO alpha-2 code for the full profile (e.g. "DE"). - `limit` (number): Return at most this many countries in a ranking. - `metric` (string): Ranking metric when neither `country` nor `countries` is given. - `sort` (string): Ranking direction. Ignored otherwise. Output parameters: - `attribution` (object) - `caveats` (array) - `metric` (string) - `ok` (boolean) - `replacement` (object) - `retirementAge` (object) - `rows` (array) - `statutoryRetirementAge` (object) ### `find_retirement_visas` (~391 tokens) Retirement and passive-income visas, matched to a monthly income Which of the 41 spine countries would let a retiree in, and on what terms — a hand-curated register of retiree, passive-income and deposit routes (kind), each with minimum age, monthly income or deposit threshold in its own currency, duration, renewability, a path-to-residency flag, the official page it was read from and the check date; an entry read from secondary reporting because the official site could not be fetched is marked `confidence: "secondary"`, and countries with no route are `kind: "none"` with a line saying why. Give `income` (monthly) with optional `currency`, `age` and `savings` for a match: routes the income reaches, near misses within 25%, deposit routes (judged only when `savings` is given — a pension says nothing about a fixed deposit), routes with no published figure, exclusions with the reason, and the countries with no route. Or `country` for one entry, `countries` for a side-by-side. Thresholds are compared in USD at the site's FX snapshot; dependants raise them. Published rules and modelled figures on the dates shown, for information only — entitlements depend on a person's record and consulates decide; every response says so first. Input parameters: - `age` (number): Applicant age; routes with a higher minimum are excluded. - `countries` (string): Comma-separated ISO alpha-2 codes, two or more, for a side-by-side. - `country` (string): One ISO alpha-2 code for a single entry. - `currency` (string): ISO 4217 code of `income` and `savings` (e.g. "GBP"). - `income` (number): Monthly income to match, in `currency`. - `savings` (number): Lump sum available, in `currency`; unlocks deposit routes. Output parameters: - `attribution` (object) - `caveats` (array) - `depositRoutes` (array) - `kind` (string) - `nearMiss` (array) - `none` (array) - `ok` (boolean) - `qualify` (array) - `rows` (array) ### `retire_on_budget` (~243 tokens) How far a monthly pension goes, country by country Does a fixed monthly pension cover a month of local essentials. Given `amount` and `currency` (default GBP 1,500), returns every spine country ranked by coverage — the pension in USD divided by the country's MODELLED monthly essentials (the site's price basket scaled by the PPP price level, checked against real series and wrong both ways) — with the essentials figure, healthcare out-of-pocket per person per month (World Bank / WHO), life expectancy (World Bank) and the kind of retiree visa the country offers beside each row. Coverage is an ordering, not a budget; the pension is taken as fixed at the site's FX snapshot. Different from `pension_abroad`, which re-prices a pension against the origin's price level to say what it is worth relative to home. Published rules and modelled figures on the dates shown, for information only — entitlements depend on a person's record and consulates decide; every response says so first. Input parameters: - `amount` (number): Monthly pension, in `currency`. - `currency` (string): ISO 4217 code of `amount`. - `limit` (number): Return at most this many countries. Output parameters: - `attribution` (object) - `caveats` (array) - `ok` (boolean) - `pension` (object) - `rows` (array) ### `compare_effective_tax` (~268 tokens) Effective corporate tax rate beside the statutory rate, by country What a company's investment is actually taxed at: OECD's forward-looking effective average tax rate (EATR), effective marginal rate (EMTR) and cost of capital for a composite investment under fixed interest and inflation, beside the statutory combined corporate rate from the Rules hub, with the gap in points — what allowances and incentives are worth. `country` for one, `countries` for a side-by-side, neither ranks by `metric` (eatr default; emtr, costOfCapital, gap), lowest first unless `sort=highest`. 39 of the 41 spine countries are modelled; the Philippines and the UAE keep their statutory rate with a caveat and never a borrowed effective rate. A property of the tax system, not any company's bill. Published rates and facts on the dates shown, for information only — not tax, legal or immigration advice; every response says so first. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more, for a side-by-side. - `country` (string): One ISO alpha-2 code. - `limit` (number): Return at most this many countries. - `metric` (string): Ranking metric when neither `country` nor `countries` is given. - `sort` (string) Output parameters: - `attribution` (object) - `caveats` (array) - `effective` (object) - `gapPct` (number) - `metric` (string) - `ok` (boolean) - `rows` (array) - `statutoryCombinedPct` (number) ### `withholding_tax` (~285 tokens) Withholding tax on payments abroad, standard and treaty rate per corridor What a company in `from` keeps back when it pays dividends, interest, royalties or technical fees to a recipient in `to`: OECD's standard withholding rates for the payer's country and, for the pair, the treaty rate where OECD publishes one, with the saving in points per income type. Direction is the payer's, never reversed; `from` equal to `to` is rejected. A pair with no published treaty rate reports the standard rate and says so — that is not the same as no treaty existing, and the treaty figure is the maximum the treaty allows (a residence certificate is usually needed). Without `from`, ranks all 41 by standard rate on `type` (dividends default). Standard rates for 41 of 41; treaty rates for about 1,300 payer–recipient pairs within the spine (the Philippines never appears as a payer). Published rates and facts on the dates shown, for information only — not tax, legal or immigration advice; every response says so first. Input parameters: - `from` (string): ISO alpha-2 code of the paying company's country. - `limit` (number) - `sort` (string) - `to` (string): ISO alpha-2 code of the recipient's country; unlocks the treaty rate. - `type` (string): Income type to rank by when `from` is not given. Output parameters: - `attribution` (object) - `caveats` (array) - `lines` (array) - `ok` (boolean) - `rows` (array) - `standard` (object) - `to` (object) - `treaty` (object) ### `compare_business_entry` (~304 tokens) Days and cost to register a company, and the new-company rate, by country How long and how much it takes to register a limited company: World Bank B-READY's calendar days and cost to complete company entry for a domestic and a foreign-owned company (cost in local currency with a USD conversion at the site's FX snapshot), whether paid-in minimum capital is required, whether sectors are closed to full private ownership, whether a national-employee quota applies, and whether registration is online with an e-signature, fees payable online and a unique business id assigned — plus the World Bank Entrepreneurship Database's new limited-liability companies per 1,000 working-age adults. B-READY is a rotating cohort: 22 of the 41 spine countries are in the 2025 edition, and a country outside it answers with the new-company rate alone and a caveat naming the gap; a country with neither is NO_DATA. `country` for one, `countries` to compare, neither ranks by `metric` (daysDomestic default; costDomesticUsd, daysForeign, costForeignUsd, density). Published rates and facts on the dates shown, for information only — not tax, legal or immigration advice; every response says so first. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more. - `country` (string): One ISO alpha-2 code. - `limit` (number) - `metric` (string) - `sort` (string): Default lowest for days and cost, highest for density. Output parameters: - `activity` (object) - `attribution` (object) - `caveats` (array) - `entry` (object) - `metric` (string) - `ok` (boolean) - `rows` (array) ### `find_founder_visas` (~350 tokens) Founder, start-up and investor visas, matched to available funds Which of the 41 spine countries would let a founder in, and on what terms — a hand-curated register of founder/start-up routes, investor routes and `kind: "none"` entries, each with its one published money figure and what it is for (`minFunds.basis`: an investment to make, subsistence funds to hold, or funding raised from investors), whether an endorsing body must approve the project, duration, renewability, a path-to-residency flag, the official page it was read from and the check date; an entry read from secondary reporting because the official site could not be fetched is marked `confidence: "secondary"`. Give `funds` with optional `currency` for a match: routes the funds reach, near misses within 25%, routes needing money raised from investors rather than held, routes with no published figure, exclusions with the reason, and the countries with no route. Or `country` for one entry, `countries` for a side-by-side. Thresholds are compared in USD at the site's FX snapshot; endorsement and business-plan tests are not modelled, and programmes pause and reprice. Published rates and facts on the dates shown, for information only — not tax, legal or immigration advice; every response says so first. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more, for a side-by-side. - `country` (string): One ISO alpha-2 code for a single entry. - `currency` (string): ISO 4217 code of `funds` (e.g. "GBP"). - `funds` (number): Funds available to commit, in `currency`. Output parameters: - `attribution` (object) - `caveats` (array) - `fundingRaised` (array) - `kind` (string) - `nearMiss` (array) - `none` (array) - `ok` (boolean) - `qualify` (array) - `rows` (array) ### `compare_health_capacity` (~243 tokens) Doctors, nurses and hospital beds per 10,000, by country How much health system there is per person: medical doctors and nursing and midwifery personnel per 10,000 (WHO Global Health Observatory) and hospital beds per 10,000 (WHO, or OECD where newer — the row names its source), beside the UHC service-coverage index. Density is not access: it says how much exists, not whether a visitor can use it or what it costs. Newest observation at the site's vintage floor; 36 of the 41 spine countries carry doctors and nurses, 31 beds; an older figure is absent, never served as current. `country` for one, `countries` to compare, neither ranks by `metric` (doctors default; nurses, beds), highest first unless `sort=lowest`. Published figures, for information only; every response says so first. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more, for a side-by-side. - `country` (string): One ISO alpha-2 code. - `limit` (number): Return at most this many countries. - `metric` (string) - `sort` (string) Output parameters: - `attribution` (object) - `bedsPer10k` (object) - `caveats` (array) - `doctorsPer10k` (object) - `metric` (string) - `nursesPer10k` (object) - `ok` (boolean) - `rows` (array) - `uhcIndex` (number) ### `compare_childcare_cost` (~256 tokens) Net childcare cost for a couple and a single parent, by country What full-time centre-based childcare costs a family after benefits and tax, from OECD's tax-benefit model (two children aged 2 and 3, no social assistance or rent supplement): a couple both earning 100% of the average wage and a single parent earning 67%, each as a share of household disposable income, of the average wage, and in local currency per year. OECD members only — 22 of the 41 spine countries at the vintage floor; the rest return NO_DATA naming the gap, never a modelled figure. `country` for one, `countries` to compare, neither ranks by `household` (couple default; singleParent) and `unit` (pctDisposableIncome default; pctAverageWage), cheapest first unless `sort=highest`. Published cost figures, for information only; every response says so first. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more, for a side-by-side. - `country` (string): One ISO alpha-2 code. - `household` (string) - `limit` (number): Return at most this many countries. - `sort` (string) - `unit` (string) Output parameters: - `attribution` (object) - `caveats` (array) - `couple` (object) - `ok` (boolean) - `rows` (array) - `singleParent` (object) - `year` (number) ### `compare_food_cost` (~370 tokens) Measured cost of a healthy diet, by country, and how far modelled grocery prices are from it What eating a healthy diet costs per person per day, MEASURED: the World Bank's Food Prices for Nutrition series (the cheapest locally available foods that meet a healthy-diet standard, from International Comparison Program retail prices), 165 of the 186 spine countries. Returns the cost in local currency (dailyLocal) and USD (dailyUsd, monthlyUsd), the share of people who cannot afford it (unaffordable.pct, on its own year), the diet's split across six food groups as percentage shares for the single year published, and how this site's MODELLED grocery prices compare: vsUsMeasured and vsUsModelled are each a multiple of the US, and modelError = modelled ÷ measured (below 1, the model understates food there). A ranking also returns modelErrorSummary for the whole site. Where the local figure does not convert consistently at the site's exchange rate (a parallel market, a recent devaluation, a currency change) the dollar figure and the comparison are withheld and a caveat says why; a country with no survey is NO_DATA, never zero. It is a least-cost diet, not what a typical or expat household spends, and it feeds no total. `country` for one, `countries` to compare, neither ranks by `metric` (dailyUsd default, or unaffordablePct), cheapest first unless `sort=highest`. Published cost figures, for information only; every response says so first. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more, for a side-by-side. - `country` (string): One ISO alpha-2 code. - `limit` (number): Maximum rows in a ranking. - `metric` (string) - `sort` (string) Output parameters: - `attribution` (object) - `caveats` (array) - `dailyLocal` (number) - `dailyUsd` (number) - `excluded` (array) - `modelError` (number) - `modelErrorSummary` (object) - `ok` (boolean) - `rows` (array) ### `compare_trade_customs` (~244 tokens) Days to clear customs and cost to comply with trade rules, by country How long goods take to clear the border and what compliance costs, from World Bank B-READY's International Trade topic (firm-survey averages in the largest business city): days for exported and imported goods to clear all border agencies, the typical cost of complying with export and import requirements as a share of goods value, and the share of firms exporting by parcel. B-READY is a rotating cohort — 22 of the 41 spine countries in the 2025 edition — and a country outside it is NO_DATA, never a borrowed figure. `country`, `countries`, or a ranking by `metric` (importDays default; exportDays, importCostPct, exportCostPct), lowest first unless `sort=highest`. Published rates and facts, for information only — not tax, legal or customs advice; every response says so first. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more, for a side-by-side. - `country` (string): One ISO alpha-2 code. - `limit` (number): Return at most this many countries. - `metric` (string) - `sort` (string) Output parameters: - `attribution` (object) - `caveats` (array) - `exportCostPct` (number) - `exportDays` (number) - `importCostPct` (number) - `importDays` (number) - `metric` (string) - `ok` (boolean) - `rows` (array) ### `compare_tax_burden` (~270 tokens) Hours on tax compliance, VAT refund weeks and the rate firms report paying, by country How much work paying tax is, from World Bank B-READY's Taxation topic (firm surveys in the largest business city): median hours spent on tax compliance a year, the share of firms filing and paying electronically, weeks to a final audit report and to a VAT refund, the share citing a too-long refund process, the share of social-security taxes in labour cost, and the effective income-based rate firms REPORT paying — which is neither the statutory rate (Rules hub) nor OECD's modelled effective rate (Business hub's tax tab), and the caveat keeps the three apart. 22 of the 41 spine countries; outside the cohort is NO_DATA. `country`, `countries`, or a ranking by `metric` (complianceHours default; vatRefundWeeks, auditWeeks, reportedEffectiveRatePct), lowest first unless `sort=highest`. Published rates and facts, for information only — not tax or legal advice; every response says so first. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more, for a side-by-side. - `country` (string): One ISO alpha-2 code. - `limit` (number): Return at most this many countries. - `metric` (string) - `sort` (string) Output parameters: - `attribution` (object) - `caveats` (array) - `complianceHours` (number) - `metric` (string) - `ok` (boolean) - `reportedEffectiveRatePct` (number) - `rows` (array) - `vatRefundWeeks` (number) ### `compare_egov` (~237 tokens) UN E-Government Development Index and its components, by country How much of government can be done online: the UN E-Government Development Index (0–1) and its three components — Online Service Index, Telecommunication Infrastructure Index, Human Capital Index — for all 41 spine countries, read from the UN E-Government Knowledgebase's country pages because no data file is published. The index measures what government provides online and the infrastructure and skills to use it, not whether a newcomer can use it (most services need a national id or residence first). The survey is biennial and the caveat names its year. `country`, `countries`, or a ranking by `index` (egdi default; osi, tii, hci), highest first unless `sort=lowest`. Published figures, for information only; every response says so first. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more, for a side-by-side. - `country` (string): One ISO alpha-2 code. - `index` (string) - `limit` (number): Return at most this many countries. - `sort` (string) Output parameters: - `attribution` (object) - `caveats` (array) - `egdi` (number) - `hci` (number) - `index` (string) - `ok` (boolean) - `osi` (number) - `rows` (array) - `tii` (number) - `year` (number) ### `property_rules` (~269 tokens) Can a foreigner buy a home there — ownership band, restrictions, transfer tax Whether a foreigner may buy residential property in each of the 41 spine countries: an ownership band (`free`, `restricted`, `leasehold-only`, `prohibited`), the main restrictions in a sentence, whether land itself may be owned, and the headline transfer tax a buyer pays with a note on surcharges and fees — a hand-curated register with the official page and check date per row; an entry read from secondary reporting because the official page could not be fetched is marked `confidence: "secondary"`. `country` for one entry, `countries` for a side-by-side; without either, all 41 ranked free-first then by transfer tax, optionally within one `band`. Rules change by decree and by region (Canada's ban runs to 2027, Australia's established-home ban to March 2027, Saudi Arabia's new law from January 2026). Published rules for information only — not legal advice; every response says so first. Input parameters: - `band` (string): Restrict the ranking to one ownership band. - `countries` (string): Comma-separated ISO alpha-2 codes, two or more, for a side-by-side. - `country` (string): One ISO alpha-2 code. - `limit` (number): Return at most this many countries. Output parameters: - `attribution` (object) - `caveats` (array) - `checkedOn` (string) - `foreignOwnership` (string) - `ok` (boolean) - `restrictions` (string) - `rows` (array) - `source` (string) - `transferTaxPct` (number) ### `employment_rules` (~341 tokens) Statutory employment floors — annual leave, notice, probation, weekly hours The minimums a job's law sets in each of the 41 spine countries: paid annual leave in days with its `basis` (`working` or `calendar`, never converted), the minimum notice an employer and an employee must give in weeks (`noticeNote` where it rises with tenure), the probation ceiling in months and the ordinary weekly-hours ceiling — a hand-curated register with the labour authority's page and check date per row, added 13 Sep 2026; an entry read from secondary reporting because the official page could not be fetched is marked `confidence: "secondary"`. A fact the law does not set is ABSENT, never zero (the US sets no paid leave, notice or hours ceiling); federal countries carry the federal floor. `country` for one entry, `countries` for a side-by-side; without either, ranks by `metric` (`minAnnualLeaveDays` default, `noticeEmployerMinWeeks`, `probationMaxMonths`, `maxWeeklyHours`), most first unless `sort=fewest`, with the countries that set no figure in `excluded`. Statutory floors only — contracts and collective agreements usually give more. For information only — not legal advice; every response says so first. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more, for a side-by-side. - `country` (string): One ISO alpha-2 code. - `limit` (number): Return at most this many countries. - `metric` (string): Ranking metric when neither country nor countries is given. - `sort` (string): Most first (default) or fewest first. Output parameters: - `attribution` (object) - `basis` (string) - `caveats` (array) - `checkedOn` (string) - `excluded` (array) - `maxWeeklyHours` (number) - `minAnnualLeaveDays` (number) - `noticeEmployeeMinWeeks` (number) - `noticeEmployerMinWeeks` (number) - `ok` (boolean) - `probationMaxMonths` (number) - `rows` (array) - `source` (string) ### `driving_rules` (~410 tokens) Road rules before you drive — side, speed limits, alcohol limit, equipment, IDP What a visitor needs before driving in each of the 41 spine countries: which side of the road, the national default urban, rural and motorway speed limits in the unit signed there (`speedUnit` km/h or mph, never converted), the blood-alcohol limit for ordinary drivers in g/100 ml (`bacNote` for novice and professional limits; 0 is a real zero-tolerance rule), equipment the law requires carried from a fixed vocabulary, and whether an International Driving Permit is needed (`idpRequired`, from GOV.UK's per-country table for a UK photocard licence; absent where that table is silent) — a hand-curated register with the authority's page and check date per row, added 13 Sep 2026; secondary-sourced entries are marked `confidence: "secondary"`. A limit not set nationally is ABSENT, never zero (Germany's motorways, US and Canadian limits). `country` for one entry, `countries` for a side-by-side; without either, ranks by `metric` (`speedMotorway` default, `speedUrban`, `speedRural`, `bacLimitPct`), highest first unless `sort=lowest`, optionally filtered by `drivesOn` and `idpRequired`; mph limits are ordered by their km/h equivalent for the sort only. For information only; conditions change — every response says so first. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more, for a side-by-side. - `country` (string): One ISO alpha-2 code. - `drivesOn` (string): Restrict the ranking to one side of the road. - `idpRequired` (string): Restrict the ranking by whether an International Driving Permit is needed. - `limit` (number): Return at most this many countries. - `metric` (string): Ranking metric when neither country nor countries is given. - `sort` (string): Highest first (default) or lowest first. Output parameters: - `attribution` (object) - `bacLimitPct` (number) - `caveats` (array) - `checkedOn` (string) - `drivesOn` (string) - `excluded` (array) - `idpRequired` (boolean) - `ok` (boolean) - `requiredEquipment` (array) - `rows` (array) - `source` (string) - `speedMotorway` (number) - `speedRural` (number) - `speedUnit` (string) - `speedUrban` (number) ### `find_student_visas` (~286 tokens) Student visas matched to the funds a student can show Student-visa routes for the 41 spine countries — a hand-curated register of the funds a student must show (amount, currency, per month or year), the work hours allowed in term, the post-study stay in months, a path-to-residency flag, the official page it was read from and the check date; secondary-sourced entries are marked, and countries with no route are `kind: "none"`. No tuition figures: no citable structured source exists. Give `funds` with optional `currency` and `per` (year default; month) for a match: routes the funds reach, near misses within 25%, routes with no published figure, exclusions with the reason, and the countries with no route. Or `country` for one entry, `countries` for a side-by-side. Funds are compared per year in USD at the site's FX snapshot; figures move each academic year. Published rules for information only — not immigration advice; every response says so first. Input parameters: - `countries` (string): Comma-separated ISO alpha-2 codes, two or more, for a side-by-side. - `country` (string): One ISO alpha-2 code. - `currency` (string) - `funds` (number): Living funds the student can show, in `currency`, for `per`. - `per` (string) Output parameters: - `attribution` (object) - `caveats` (array) - `kind` (string) - `nearMiss` (array) - `none` (array) - `ok` (boolean) - `qualify` (array) - `rows` (array) - `unpublished` (array) ### `find_country_fit` (~376 tokens) Which country fits me — ranked by your own weights Ranks countries by a fit score that blends the livability composite (internet speed, safety, healthcare access, air quality, English — measured, each with its own source year) with how far a monthly budget in USD goes against modelled essentials, using 0–10 weights per dimension (`cost`, `internet`, `safety`, `healthcare`, `air`, `english`; default 5 each). Region, budget and visa are FILTERS, never weights: `region` (a World Bank region name), `withinBudget` (default true; false keeps countries the budget does not cover), `needsVisa` (an eligible digital nomad visa on the budget). Each row carries the fit, quality and cost scores, modelled essentials, budget ratio and the nomad visa matched. A ranking of published figures against the weights given, not a recommendation; every response says so first. Input parameters: - `air` (number): Weight 0–10 for air (default 5). - `budget` (number): Monthly budget in USD. - `cost` (number): Weight 0–10 for cost (default 5). - `english` (number): Weight 0–10 for english (default 5). - `healthcare` (number): Weight 0–10 for healthcare (default 5). - `internet` (number): Weight 0–10 for internet (default 5). - `limit` (number): Maximum rows (default 10). - `needsVisa` (boolean): Keep only countries with a digital nomad visa the budget meets. - `region` (string): Keep only countries in this World Bank region. - `safety` (number): Weight 0–10 for safety (default 5). - `withinBudget` (boolean): Keep only countries whose modelled essentials fit inside the budget. Output parameters: - `attribution` (object) - `caveats` (array) - `excluded` (object) - `inputs` (object) - `ok` (boolean) - `rows` (array) ### `country_dossier` (~596 tokens) One country, every hub: the headline figures The three to six headline figures each hub owns for one country, in one call: price level vs the US and datasets covering it (compare); monthly essentials, health out-of-pocket spending and — with `income` — what is left after costs (costs); modelled median and minimum wage, ILOSTAT measured median, pension replacement rate (earn); safety and healthcare-access scores, rule of law, disaster risk, comfortable days, tax-residency threshold, passport rank, nomad-visa programmes (move); public holidays, latency and download speed, employer contribution loading, weekly hours, paid maternity leave (work); inflation, policy rate, banked adults, remittance corridors surveyed (money); combined corporate tax rate, top income-tax rate and where it starts, standard VAT rate (absent for the US — state sales taxes are not a 0% VAT), the Women, Business and the Law index and retirement ages for men and women (rules); FCDO and US State Department advice levels, the yellow-fever certificate position, homicide rate and PM2.5 exposure with their years (safety); retirement ages for men and women (OECD, else WBL), the net replacement rate, life expectancy and the kind of retiree visa (retire). Effective average corporate tax rate, standard withholding on dividends, days to register a company (absent outside the B-READY cohort) and the kind of founder visa (business). Since 5 Sep 2026 the sections also carry doctors per 10,000 (safety), the couple's net childcare cost as a share of income (costs), days for imports to clear customs and hours on tax compliance (business), the E-Government index (work), and the foreign-ownership band for property and the kind of student visa (move) — each absent where its source does not cover the country. Every figure carries the id of the source that produced it, `tools` names the MCP tool to call for detail per section, and `sources` lists only what was actually included. A section or figure the site has no data for is ABSENT, never null — Thailand has n… Input parameters: - `country` (string, required): ISO alpha-2 code (e.g. "PT"). - `income` (number): Gross monthly income in USD. Optional; unlocks the total-cost figures. Output parameters: - `attribution` (object) - `country` (object) - `ok` (boolean) - `sections` (object) - `sources` (array) - `tools` (object) - `updated` (string) ### `list_datasets` (~322 tokens) Every dataset on the site, and when it last changed The registry of every dataset behind the site's figures — price levels, wages, housing, healthcare, holidays, working time, inflation, remittance prices, and the rest. Each carries `updated`, the date it was last REBUILT (a fetcher ran, or a person reviewed it), and — where the dataset knows it — `vintage`, what the figures themselves describe: a year, an edition, a quarter, or the source's own publication date. THE TWO ARE DIFFERENT AND OFTEN YEARS APART: exchange rates rebuilt in 2026 are 2025 annual averages, and city rents rebuilt in 2026 are from 2023. Judge how current a figure is by `vintage` and fall back to `updated` only where `vintage` is absent, which means the dataset has none of its own — a curated register's vintage is the day a person read the authority's page. Also says whether it is FETCHED by a script or REVIEWED by a person, the page it appears on, and the refresh script. With `since` (YYYY-MM-DD), `changed` lists only the datasets rebuilt strictly after that date, so an agent that cached a figure can decide what to re-fetch instead of re-fetching everything. Pure: `since` comes from the caller. Carries no figures and therefore no attribution block; each dataset's provenance travels with the tool that serves it. Input parameters: - `since` (string): Return only datasets changed strictly after this date (YYYY-MM-DD). Omit for all. Output parameters: - `changed` (array) - `changedCount` (number) - `count` (number) - `datasets` (array) - `newest` (string|null) - `ok` (boolean) - `since` (string|null) ### `compare_climate` (~313 tokens) Compare climate normals and comfortable days Monthly climate normals from ERA5 reanalysis (2015-2024) via Open-Meteo: mean daily maximum and minimum temperature, average monthly rainfall, and a count of comfortable days. Pass `city` (slug) or `country` (alpha-2) for one place's twelve months — a country answers through a named representative city, because a country does not have one climate. Otherwise cities are ranked by comfortable days, for a single `month` (1-12) or across the year; the seasonal question is the one people actually have, and an annual total hides it. A comfortable day has a maximum between 18 and 27 C with under 1 mm of rain — a STATED PREFERENCE, not an objective measure: it is temperate-biased and scores hot destinations low on terms they did not agree to, so mean maximum and rainfall travel with every row for callers who want to rank on those instead. ERA5 is reanalysis rather than station observation, which makes it complete and comparable across countries. Input parameters: - `city` (string): City slug (e.g. "lisbon"). Returns that city's twelve months. - `country` (string): ISO alpha-2. Returns the representative city's months, naming which city. - `limit` (number): Return at most this many cities. - `minComfortDays` (number): Filter: minimum comfortable days in the ranked window. - `month` (number): Rank a single month. Omit to rank on the whole year. Output parameters: - `city` (object) - `month` (number|null) - `ok` (boolean) - `rows` (array) - `updated` (string) ### `compare_country_risk` (~255 tokens) Compare country risk (INFORM, EC JRC) INFORM Risk from the European Commission's Joint Research Centre: a 0-10 composite where HIGHER IS WORSE, plus the three dimensions behind it — hazard (exposure to earthquakes, floods, cyclones, drought, conflict), vulnerability (how exposed the population is) and copingCapacity (whether institutions and infrastructure could respond). The last is usually the one that matters when choosing where to live: two countries can face the same earthquake and have completely different outcomes. Safest-first by default. Covers 191 countries, far wider than this site's country pages. A country INFORM does not score is ABSENT, never zero — zero is the safest possible value and would rank an unassessed country first. Deliberately not folded into any other ranking here: risk flags, it never silently reweights a cost or livability score. Input parameters: - `country` (string): ISO alpha-2 for a single country's scores. - `descending` (boolean): True lists the most dangerous first. - `limit` (number): Return at most this many countries. - `maxRisk` (number): Keep only countries at or below this composite score. - `sort` (string): Dimension to sort by. maxRisk always filters on the composite. Output parameters: - `entry` (object) - `ok` (boolean) - `release` (string) - `rows` (array) - `updated` (string) ### `arrival_cost` (~347 tokens) What It Costs to Arrive The cash needed to ARRIVE somewhere, as opposed to what a month costs once you live there: the entry fee for a given passport, the first month's rent, the deposit, and a long-stay visa fee where one has been verified against the issuing government. READ `complete` BEFORE `totalUsd`. Rent covers a subset of the country spine and most visa fees could not be verified, so the total is a FLOOR whenever `complete` is false, and `missing` names every component that could not be priced. The total can legitimately be 0 - visa-free entry into a country with no rent figure sums to nothing known - which is honest arithmetic and a false headline. `depositMonths` is a STATED ASSUMPTION, not a measurement: no free global dataset of deposit norms exists. The entry route itself is derived exactly from passport mobility data, so 'for this passport, this destination is visa-free' is always answered even when no fee is. Input parameters: - `depositMonths` (number): Months of rent held as a deposit. Defaults to the site's stated assumption of 1; override it if you know the market. - `destination` (string, required): ISO alpha-2 of the destination. Must be in the PPP country registry; some countries with a verified nomad-visa fee (e.g. BB) are not and appear only in find_nomad_visas. - `income` (number): Monthly remote income in USD. Optional; enables housing's income-relative view but is not needed for the arrival total. - `passport` (string, required): ISO alpha-2 of the traveller's passport, e.g. 'US'. Determines the entry route and therefore the fee. Output parameters: - `attribution` (object) - `excluded` (number) - `focus` (object) - `ok` (boolean) - `rows` (array) - `updated` (string) ## Diagnostics Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/com-geo-parity-geo-parity/api-mcp#diagnostics ## Score history - 2026-09-21: 87 - 2026-09-20: 86 - 2026-09-19: 86 - 2026-09-18: 84 - 2026-09-17: 83 - 2026-09-16: 83 - 2026-09-15: 83 - 2026-09-14: 82 - 2026-09-13: 82 - 2026-09-12: 81 - 2026-09-11: 81 - 2026-09-10: 78 - 2026-09-09: 77 - 2026-09-08: 77 - 2026-09-07: 76 - 2026-09-06: 76 - 2026-09-05: 75 - 2026-09-04: 75 - 2026-09-03: 75 - 2026-09-02: 76 - 2026-09-01: 75 - 2026-08-31: 75 - 2026-08-30: 74 - 2026-08-29: 74 - 2026-08-28: 73 - 2026-08-27: 73 - 2026-08-26: 66 ## Common questions ### What is the Geo-Parity MCP server? Geo-Parity is an MCP server listed in the public MCP registry as com.geo-parity/geo-parity. Cost of living, fair global pay, healthcare costs, nomad visas and tax residency. Free, no key. This page covers its hosted endpoint (https://geo-parity.com/api/mcp). ### Is the Geo-Parity MCP server safe to use? Geo-Parity scores 87 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm. ### What tools does the Geo-Parity MCP server expose? Geo-Parity exposes 62 tools: compare_total_cost, calculate_ppp, localize_salary, calculate_wage_bands, compare_occupation_pay, and 57 more. Their descriptions and schemas cost roughly 17,916 tokens of context every time the server is loaded. ### Does the Geo-Parity MCP server require authentication? No. We connected to Geo-Parity without credentials and it answered, so anything it exposes is reachable by anyone who knows the address. ### Is the Geo-Parity MCP server still maintained? Geo-Parity is still listed as active in the MCP registry. We last reached this channel on 21 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced. ## Links - Remote endpoint: https://geo-parity.com/api/mcp - Website: https://geo-parity.com/ - Changelog RSS feed: https://verifymcp.io/servers/com-geo-parity-geo-parity/api-mcp.xml - Changelog JSON feed: https://verifymcp.io/servers/com-geo-parity-geo-parity/api-mcp.json - HTML version of this page: https://verifymcp.io/servers/com-geo-parity-geo-parity/api-mcp