RateSonar MCP Server

Recent change-events, newest first: a rate moved, a product was added, or a product was withdrawn. Each event has the lender, the product's name and category, and field-level deltas (e.g. rate [6.09, 5.89], bps -20). Filter by institutionId, category, kind and direction. Returns `limit` events (default 50, at most 200); `truncated` says more match. This is a diff of RateSonar's consecutive captured observations.

Exact CDR product-reference detail for one institution and product: every published lending or deposit rate entry and tier, with fees, features, eligibility, constraints, source quality flags, the date the version was recorded and the lender's own links. asOf returns the version current on that date. Detail exists for CDR products from 2026-06-12; it holds the conditions a headline rate leaves out.

Advertised rate history for one product (institutionId and productId). Returns each recorded rate, comparison rate and fees between two dates, with the date each was recorded. RateSonar's record starts 2026-06-12. Pass grid:true to also get `grid`, one point per day: status 'observed' on a recorded day; 'carried-forward' with sourceDate and fill {from, through, ageDays, reason:'no-record'} on a day RateSonar has no record for, carrying the last recorded rate (never the next one) until the next record. After the last record it carries only while RateSonar's gap ledger shows the product still expected but unrecorded (not withdrawn), for at most 14 days. A carried-forward value is not an observation. History is limited to the caller's plan window (free and anonymous 31 days, Enthusiast 366, Pro full); an older `from` is raised to it and the answer says windowClamped:true.

An evidence-checked AI reading of bank_position. Supports savings or term deposits against variable/fixed home loans, card purchase rates, personal, car or business loans. Returns cited claims, inference labels and dated records; partial readings name withheld sections; unavailable or rejected AI leaves the records intact. No supplied rates accepted.

Selected bank deposit and loan contracts against matching peer brands, own recorded history, RBA cash and dated funding disclosures. Pass an institutionId and product cohort. No claims about private ALM, intent or margins.

Every standard series slug the series tool accepts, with its label, kind (level or spread) and unit (% or bps). Call this before series when you do not know the slug.

Standardized rate series: named LEVEL series (hl.var.oo.pi = headline variable home loan; sav.max; td.12m; biz.loan.var; cash = RBA cash rate…) and SPREAD series in bps (spread.hl.var-cash = lending pass-through; spread.hl.var-sav.max = the margin; spread.hl.var-hl.fix.36m = variable vs 3yr fixed…), per institution per day. Last-known values are explicitly status:'carried-forward' with sourceDate and fill {from, through, ageDays, reason:'no-record'} (through = the next recorded date; null = the gap is open, or no record within 60 days after the window end). Between two records the last rate is carried until the next one. After an institution's last record it is carried only while RateSonar's gap ledger shows the product that holds that rate still expected but unrecorded (not withdrawn), for at most 14 days, then the series stops; medians and spreads use the same points, so a stale value never feeds them past 14 days. The RBA cash series publishes on business days only, so its carried points have no fill; it is carried at most 14 days after its last row. Spreads and medians are status:'derived' with sourceDates; they are not observations. Pass median:true for a market-median line. Unauthenticated access and free access are clamped to the last 31 days, Enthusiast to 366 (windowClamped:true in the response). Pro tiers query deeper history in windows of at most 367 inclusive dates; large multi-series median requests may require smaller windows.

Current comparison across institutions for a product category (home_loan | savings | term_deposit | credit_card | personal_loan | car_loan | business_loan): each product at its own most recent observation (observedDate tells you how fresh). Home loans can be filtered by loanPurpose/repaymentType/rateType — the headline home-loan rate is the owner-occupied P&I rate whose LVR band covers 80%. Term deposits take termMonths (the published rate for that term, e.g. 6 or 12) and deposits take customerSegment. Standard products only unless audience says otherwise. Returns the best `limit` rows (default 25, at most 40); `truncated` says more exist. It has no lender filter: for one lender's products use find_products. Fees are summarised (name, type, amount or rate, frequency); rate_details has every fee field.

Validation, freshness, missing families, source hashes and last-good fallback status of the automated curve pipeline. An unhealthy response is evidence of a pipeline problem, not bank risk.

Find current products by lender and words in the product name, e.g. institution "CBA" and query "standard variable". Returns institution_id and product_id for each match, with category, rate and observed_date. A product is the pair (institution_id, product_id): pass both to rate_history or rate_details. Use compare_rates instead to rank lenders for a category.

RBA curve nodes for an exact date or the latest publication: cash and bond yields as published, and the RBA's fitted zero, forward and discount curves (each point says whether it is observed or derived), with optional exact comparison date, instrument details, source digest and convention. Eight distinct cash/government families; no intraday OIS, BBSW, swap or futures feed. Zero/forward fitted data publish monthly.