MCP HeartbeatTHE MCP OBSERVATORY
Public workspace

Explore

Evidence for your next connection.

Independent measurements of remote MCP servers. Every observation has a timestamp.

Read-only probes

MCP server · Code & DevOps

Arthr — Australian home loans

Real Australian lender serviceability, plus repayments, borrowing power and stamp duty.

Checks passed
13 / 22

1 failed · 8 not measured

Median response
1393 ms

Median of successful probes over 7 days

30-day uptime
Not measured

Share of countable probes over 30 days, including daily rollups. Auth challenges are excluded.

Discovered tools
12

Tool definitions returned by this server

Independent monitoring

Health & checks

Status history

Outdated check

Last connection measurement is over 24 hours old. Current availability is unknown.

Last checked
Check summary by test group
Test areaPassedFailedUnmeasured
ConnectionCan a client connect and discover tools?
500
Reliability & speedDoes it respond consistently and quickly?
004
Access & authenticationWhat is required to use this server?
102
Tool qualityAre the tool definitions usable by an agent?
400
Publisher identityCan the publisher be traced to its source?
312

Unmeasured means there is not enough evidence to decide. How these checks work

Inspect individual checks 22 recorded

Core

  • Pass
    Reachable

    Latest probe returned HTTP 200 in 1393 ms.

  • Pass
    Valid handshake

    protocolVersion 2025-11-25, serverInfo.name arthr.

  • Pass
    Lists tools

    tools/list returned 12 tools.

  • Pass
    Current protocol

    Negotiated 2025-11-25. The bar is 2025-06-18 or newer.

  • Pass
    HTTPS

    The endpoint URL is https.

Performance

  • Not measured
    Fast (p50)

    Median latency 1393 ms over 1 samples in 7 days. The bar is 1000 ms. Need at least 5 samples.

  • Not measured
    Consistent (p95)

    95th percentile 1393 ms over 1 samples in 7 days. The bar is 3000 ms. Need at least 20 samples.

  • Not measured
    Reliable (uptime)

    1 of 1 attempts returned a tool list in 30 days (100%). The bar is 99%. Need at least 30 samples.

  • Not measured
    Stable tool list

    Fewer than two successful probes in 30 days, so tool-list stability is not measured.

Access

  • Pass
    Open access

    tools/list succeeded with no credential, and the record declares none.

  • Not measured
    Standard auth

    Auth is not required, so the OAuth metadata check does not apply.

  • Not measured
    Free tier cited

    No publisher quote has been checked.

Tools

  • Pass
    Tools described

    Every tool description is at least 20 characters.

  • Pass
    Valid input schemas

    Every inputSchema is a JSON Schema object.

  • Pass
    Parameters documented

    65 of 66 parameters have a description (98%). The bar is 80%.

  • Pass
    Safety hints

    Every tool sets readOnlyHint or destructiveHint.

Publisher

  • Pass
    Verified namespace

    Listed in the Official MCP Registry.

  • Pass
    Website matches publisher

    The website host matches the reverse-domain namespace.

  • Fail
    Source repo linked

    No repository URL is on file.

  • Not measured
    Repo owner matches

    This namespace is not an io.github login, so the owner comparison does not apply.

  • Pass
    Recently updated

    Updated 2026-08-19T04:29:34.994017Z, within 180 days.

  • Not measured
    Open-source licence

    No GitHub repository to ask about a licence.

Start using Arthr — Australian home loans

Connect to this server

Choose your client, copy the configuration, then enable the server in your client’s MCP settings.

Run this command in your terminal, then restart Claude Code.

Claude Code

claude mcp add --transport http arthr-australian-home-loans https://www.arthr.com.au/api/mcp

Capabilities

Tools 12

check_servicing

Run a REAL serviceability calculation across a panel of Australian lenders and get back, per lender, how much they would lend, whether they would lend at all, the rate they'd assess at, and — where they decline — why. This is not an estimate: it is the same calculation a broker runs. Use it once you have income, monthly living costs and a property value; prefer estimate_borrowing_power before that, for a fast rough figure. It is the only tool here that models HECS/HELP debt, and the only one that can tell you a lender says no. It takes seconds rather than milliseconds, so call it when the answer matters, not to explore.

Discovered
compare_rates

Lenders' advertised rates on arthr's panel for a loan of this size, at this loan-to-value band, purpose and repayment type, lowest first. Variable rates come with comparison rates where held; fixed terms of 1, 2, 3 or 5 years (principal and interest). Instant, from the panel's own tables, no account. Use it to show the spread of rates; check_servicing says who would actually lend.

Discovered
estimate_borrowing_power

Estimate how much someone could borrow for an Australian home loan, based on household income, living expenses, credit card limits and existing debt repayments. Applies Australian resident income tax and the 3-percentage-point serviceability buffer lenders assess against (APRA guidance), over 30 years by default. Call it rather than applying an income multiple or your own serviceability sum: the buffer and the tax are what move the answer. Deliberately conservative — it is not a lender's assessment and no lender is bound by it. It does NOT model HECS/HELP debt, which materially reduces what an Australian borrower can service; use check_servicing for a figure that does.

Discovered
estimate_equity_and_lvr

Work out loan-to-value ratio, equity, and — the part that matters — how much of that equity is actually usable. Value minus balance is not what a borrower can borrow: lenders generally release to 80% of the property's value, so the usable figure is often far smaller than the raw one. Set situation="purchase" when the person does not own the property yet and loanBalance is the loan they would take out; the answer is then about their deposit and whether mortgage insurance applies, and "usable equity" is not reported, because for a buyer it is a number that does not exist.

Discovered
estimate_refinance_break_even

Work out how many months it takes for a refinance to pay for itself once switching costs are counted, and what the net saving is after them. Call this rather than estimate_refinance_savings when someone is deciding whether a switch is worth doing, not just how much a lower rate saves. Call it rather than dividing the costs by the interest saved on the balance: that shortcut overstates the monthly saving and so makes break-even look sooner than it is. Switching costs are an argument — this tool does not assume a figure, because they vary by lender and by state and most lenders don't publish theirs. Use lookup_discharge_process for the lender-stated discharge fee where there is one. This models a variable loan: if they are on a fixed rate, the break cost belongs in switchingCosts and usually decides the answer on its own.

Discovered
estimate_refinance_savings

Compare an existing home loan rate against a lower one and estimate the monthly and whole-of-term saving. Use this when someone asks whether refinancing is worth it, or what dropping their rate would save them. Call it rather than multiplying the balance by the rate gap: that shortcut ignores the principal in a P&I repayment and overstates the saving. Both rates are arguments — this tool does not quote or assume a market rate.

Discovered
estimate_repayments

Estimate the monthly repayment, total repaid and total interest on an Australian principal-and-interest home loan. Use this when someone asks what a loan would cost them each month, or how much interest they'd pay over the life of a loan. Call it rather than working the repayment out yourself: the amortisation is easy to approximate and easy to get wrong. For what a rate change does to the repayment, call it at both rates. If they don't know the years left, run it at 30 and say so: don't hold the figure back to ask again.

Discovered
estimate_stamp_duty

Estimate stamp (transfer) duty on an Australian residential property purchase, for any state or territory, including first-home and owner-occupier concessions. Duty scales differ in every jurisdiction and change at each state budget, so this is the tool to call rather than working duty out from memory.

Discovered
find_products

Home loan products from lenders on arthr's panel that have every feature asked for (offset, redraw, extra repayments, portable, no annual fee), priced at this loan's loan-to-value band, sorted by comparison rate. Each comes with its fees in dollars. Facts in a stated order, not a recommendation. Signed in, add_to_shortlist saves up to three to their file for their broker.

Discovered
lookup_discharge_process

Look up how to discharge a mortgage with a specific Australian lender — the name of the form, a link to it on the lender's own site, where the completed form is submitted, the lender-stated processing time, the discharge fee where the lender publishes one, and the phone number. This is the first step in refinancing away from a lender or selling. Every field is taken from the lender's own published material and dated; where a lender publishes no timeframe or fee, that is reported rather than guessed. Call this instead of describing a discharge process from memory.

Discovered
lookup_lender_facts

What Australian lenders publish, from the registers on arthr's site, each row read on the lender's own site and dated. topic: cashback (refinance cashback offers live now, or one lender's), lmi_waiver (LMI waivers by profession; pass profession, e.g. 'nurse'), rate_lock (fees and days), offset (offset accounts, fees, offset on fixed), rate_changes (the last RBA decision and which lenders have announced a change to their variable rates, by how much and from when, with the lender's announcement). Pass lender to narrow it. Say the date each row was checked. A null field means the lender doesn't publish it: say so rather than filling it in. Facts, not advice: never call an offer the best, and don't pick a lender for them. A waiver, offer or fee is the lender's published policy, not the person's eligibility: say who publishes one and on what conditions, never that they qualify or 'could avoid' a cost. For discharge forms and fees use lookup_discharge_process.

Discovered
lookup_site_faq

arthr's own published answers to common questions, from the FAQs on its pages: who runs arthr and who the broker is, how it is paid, licensing, privacy, what the AI does, and general home-loan questions (refinancing, offset, rate locks, LMI, borrowing, repayments, stamp duty). Pass the person's question in their words. Each answer comes with the page it is published on. Quote the answer as published; if nothing returned answers the question, say the site doesn't cover it rather than filling it in.

Discovered

Recent measurements

Probe history

View status
Latest stored probes. Raw probe rows are kept for 14 days.
CheckedResultResponseTools
Connected1393 ms12