MCP SERVERIsraeli price data straight from the Central Bureau of Statistics: find the index for bread or housing, pull its history, and run the official linkage calculator on a shekel amount.
LABSCO SUMMARY
The path to any answer is the same: search the catalogue for a keyword, take the index code, then ask for its series or hand it to the calculator. What makes the output trustworthy is that it carries the bureau's own qualifications rather than flattening them — provisional housing values are marked as provisional, and every code reports the period it actually covers, so a short series is never mistaken for a long one.
A read-only client for the CBS price-index API. It walks the catalogue from chapters down to individual index codes, returns the time series behind any of them, and runs the bureau's own inflation-adjustment calculation.
No API key — CBS data is free and public, and the server only reads it over HTTPS. It runs over stdio as npx -y @reuvenorg/israel-statistics-mcp, with a Docker image as an alternative. Date formats differ by tool and matter: mm-yyyy for an index series, yyyy-mm-dd for the calculator, and compact yyyymm for the headline range, which cannot start earlier than 199701. History depth varies by index — the general CPI runs back decades while others do not. Housing figures update every two months and the last three published values are provisional, which the server flags on its own. MIT.
One command — npx -y @reuvenorg/israel-statistics-mcp
USE CASES
Pick the index that applies, then run the official linkage calculation between the two dates rather than approximating from a headline inflation figure.
Search the catalogue by keyword, or walk down from chapter to subject to code, and read each code's coverage and update frequency before using it.
Pull the series for the code over a month range, with chaining coefficients when the comparison crosses a base change.
Read the headline indices with their latest monthly change, or the same set across a range of months when the question is about direction.
Filter the current index tree to the housing chapter, and take the provisional marking on recent values seriously — they are revised.
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
get_all_indicesThe complete current-month index tree, optionally filtered to one chapter (e.g. chapter='aa' for housing — the summary then carries the bi-monthly/provisional caveat). Large payload: prefer chapter filtering. The indices field mirrors the raw CBS XML structure (values arrive as string arrays).
get_catalog_chaptersList all CBS index chapters (a=Consumer Price Index, aa=Housing Market, b/ba/bb=Producer Prices, c/ca=Building Inputs, d=Road Construction, e=Agriculture, f/fa=Bus & Minibus Inputs, plus any newly added). This is the authoritative chapter list — use a chapterId with get_chapter_topics or get_all_indices.
get_chapter_topicsList the subjects/topics inside one chapter (chapterId from get_catalog_chapters, e.g. 'aa' for the housing market). Returns subjectIds for get_subject_codes; code arrays may be null at this level — that is normal, drill down via get_subject_codes.
get_index_calculatorInflation-adjust a monetary amount between two dates using a specific index (the official CBS linkage calculator). Dates: use yyyy-mm-dd (recommended, e.g. '2020-01-15'); dd-mm-yyyy is auto-corrected only when unambiguous (day>12) — otherwise it is read as the CBS mm-dd-yyyy convention. Some series (e.g. 110050) legitimately return null coefficient bounds. For housing indices, avoid toDate values inside the provisional window flagged in the summary. Use indexCode 120010 for general CPI.
get_index_dataRetrieve the historical time series for one index code (from get_subject_codes or get_index_topics). Periods use mm-yyyy (e.g. startPeriod='01-2020', endPeriod='12-2024'); use last=N for just the most recent N points; coef=true adds chaining coefficients to previous base periods. Housing indices (codes 18xxxx) are bi-monthly with provisional recent values — the summary warns when that matters.
get_index_topicsBrowse or search the full CBS catalog tree of price-index topics and codes. The best starting point for keyword discovery (e.g. searchText='bread' or 'housing'). Returns chapters with their subjects and index codes; feed a codeId into get_index_data or get_index_calculator. Hebrew is the default language — pass lang='en' for English names.
get_main_indicesCurrent snapshot of the headline CBS indices (general CPI, housing, producer prices, …) with values across base periods and the last monthly percent change. No parameters required; oldFormat=true switches to the legacy layout.
get_main_indices_by_periodHeadline indices over a month range. Dates use compact yyyymm (startDate='202401', endDate='202403'); the range must start no earlier than 199701 and endDate must not precede startDate. Results come both as a flat list and grouped by month.
get_subject_codesList the concrete index codes for one subject/topic (subjectId from get_chapter_topics or get_index_topics), optionally filtered by searchText. Each code includes its date coverage and update frequency — use codeId with get_index_data or get_index_calculator.
BEFORE YOU INSTALL
9 tools when we connected.
npx on PATH
All 4 documented clients run the server on your machine.
CONTEXT COST
Rated by the size of the 9 tool definitions that ride along on every turn, not by how many there are
INSTALLATION
npx -y @reuvenorg/israel-statistics-mcp{
"mcpServers": {
"israel-statistics": {
"command": "npx",
"args": [
"-y",
"@reuvenorg/israel-statistics-mcp"
]
}
}
}claude mcp add israel-statistics -- npx -y @reuvenorg/israel-statistics-mcp{
"mcpServers": {
"israel-statistics": {
"command": "docker",
"args": [
"run",
"--init",
"--rm",
"-i",
"reuvenaor/israel-statistics-mcp:latest"
]
}
}
}MCP servers
SidneyBissoli
isakskogstad
fxmacrodata
stefanoamorelli