io.github.cyanheadsPublished by GitHub account cyanheadsv0.6.0Updated Nov 3, 2026
Query 29,500+ World Bank development indicators for 200+ countries across 60+ years.
Worldbank's hosted MCP server (worldbank.caseyjhand.com); no sign-in required; Streamable HTTP; 9 tools.
No AI overview yet: these facts come from the registry entry and the server itself.
Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.
Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.
Read https://sourceweft.com/mcp/io.github.cyanheads%2Fworldbank-mcp-server/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.Add this to your client's mcpServers config.
{
"mcpServers": {
"worldbank": {
"type": "http",
"url": "https://worldbank.caseyjhand.com/mcp"
}
}
}Query 29,500+ World Bank development indicators for 200+ countries across 60+ years via MCP. STDIO or Streamable HTTP.
Public Hosted Server: https://worldbank.caseyjhand.com/mcp
World Bank Open Data: 29,500+ development indicators, poverty and inequality estimates, and the Bank's lending projects. Search indicators, query country and source-specific time series, pull poverty metrics, and search active and historical projects. No API key required. The published worldbank-guide skill (skill://worldbank-guide/SKILL.md) covers dataset selection, comparable observations, source-specific identifiers, and lending totals.
worldbank_list_topics tool1 Agriculture, 3 Economy & Growth) feed topic_id on worldbank_search_indicatorsworldbank_list_sources tool2 for World Development Indicators) feed source_id on worldbank_search_indicators, worldbank_list_countries, and worldbank_get_dataworldbank_list_countries toolEAS, SSF, … plus membership groupings such as AFE, EUU, LDC), income level (LIC, LMC, UMC, HIC), and lending type (IDX, IBD, IDB, LNX), combined by AND; up to 300 per pageinvalid_filter errorinclude_aggregates=true adds regional, income-group, and world aggregates, flagged isAggregatesource_id, returns that dataset's canonical entities as id, name, and sourceId, including subnational IDs. Global classification filters and include_aggregates=true fail as unsupported_source_filter; the default false does not discard source entitiesworldbank_get_country toolUS), ISO3 (USA), or World Bank aggregate code (EAS, HIC, WLD) — all or a list is rejected as multiple_countriescountry_not_found error with a recovery hint pointing to worldbank_list_countriesworldbank_search_indicators toolquery, topic_id, or source_id; every keyword term must match the indicator ID, name, or description, in any order; up to 100 per pageSP.DYN.LE00.IN ahead of SP.DYN.LE00.FE.IN)worldbank_get_indicator returns the wholeworldbank_get_indicator tool., _, or - — all, a list, or any other character is rejectedindicator_not_found error pointing to worldbank_search_indicatorsworldbank_get_data toolcountries as ISO2, ISO3, aggregate codes, WLD, or all alone; at most one of date_range (a year, quarter, or month, or a range of one type), mrv (the latest N periods across the countries), or mrnev (each country's own latest N non-empty values); frequency (annual, quarterly, monthly) picks the period form on series that publish several, while a date window's own form selects it; up to 1,000 observations per pageisAggregate, and value: null for a missing cell, with nullCount per page and lastUpdated, the source's data vintage; every date_range, mrv, and mrnev read is checked for another query's cached answer and re-read once, and one that fails twice is an upstream_inconsistent error rather than a served resultsourceScoped with the release, classification, sector, or counterpart area applied (dimension_value)source_id selects a dataset explicitly; discover its entity IDs with the same source on worldbank_list_countries. Source-only entities retain their opaque countryCode and an empty countryIso3; sourceId names the serving dataset when known. Unknown or unsupported sources fail as source_not_found / source_not_queryableworldbank_get_poverty toolAFG), plus PIP's aggregates: WLD, regions (SSF, AFE), income groups (LMIC/LMC), lending groups (IDX, IDB, IBD, REST), and fragility groups (FCVY, FCVN; WDI's FCS is rejected with FCVY guidance, without asserting the same membership); year as a year, all, or MRV; any poverty_line, defaulting to the applied PPP vintage's international line; up to 1,000 estimates per pageestimationType separates survey rows from gap-filled ones, whose inequality fields are null, and isAggregate flags aggregate rowsestimateType preserves actual/projection/nowcast provenance independently of estimationType; distributionType preserves micro/group/imputed classification. Both remain null when upstream supplies nonefill_gaps (default true) includes gap-filled years; welfare_type and reporting_level narrow economy results, and ppp_version picks the PPP vintageworldbank_search_projects toolquery combined by AND with countries (ISO3, ISO2, or a World Bank regional code such as 3A), region, sector, status, financial_type, and a board-approval window (approved_from/approved_to); up to 1,000 projects per page, newest approval firstcountryCodes chain into the other tools, a bad code or unparseable query is a typed invalid_country_code / invalid_query error, and an empty result names the filter that emptied itinclude_abstract (off by default) adds complete abstracts without changing the page sizesector accepts an array of names, combining them by OR and with other filters by AND. Bare and FY17 - spellings query both naming forms; codes are not aliases. Unknown names return no matches; blank names and embedded caret separators fail as invalid_sector_nameworldbank://indicator/{indicatorId} resourceapplication/json — name, description, unit, source dataset, source organization, and topicsindicatorId comes from worldbank_search_indicators; an unknown ID returns a typed not-found error with a recovery hint, while an upstream outage or timeout keeps its own classification instead of reading as a bad IDworldbank://country/{countryCode} resourceapplication/json — ISO codes, region, income level, capital, coordinatesall or a list is rejected; an unknown code returns a typed not-found error, distinct from a transient upstream failureBuilt on @cyanheads/mcp-ts-core: stdio and Streamable HTTP transports, pluggable auth (none / jwt / oauth), swappable storage (in-memory, filesystem, Supabase, Cloudflare KV/R2/D1), structured logging with optional OpenTelemetry tracing.
World Bank-specific:
searchterm parameter doesn't filternull observations and nullCount surfaced rather than silently droppedisAggregate distinguishes global country and data rows from aggregates; source-entity listings retain their own identifiers without inventing global classificationsAgent-friendly output:
worldbank_search_indicators names worldbank_list_topics for topic IDs, worldbank_get_data names worldbank_search_indicators for indicator discoveryreason codes and actionable recovery hints on every tooltotalCount, currentPage, totalPages) across all list/search/data tools, with a notice naming the pages that exist when a request runs past the endper_page up to their documented maximum; larger requests are served at that maximum with requested/served sizes and a notice. The default is 50Connect to the hosted instance, no install:
Or run it locally:
For other clients, add this to the MCP config file:
No API key required. For Docker or a source build, see Running the server.
All configuration is validated at startup via Zod schemas in src/config/server-config.ts. Environment variables:
See .env.example for the full list of optional overrides.
Requires Bun v1.4.0 or higher (or Node.js v24+).
Clone and install:
Build and run the production version:
Run checks and tests:
The image defaults to HTTP transport, stateless session mode, and logs to /var/log/worldbank-mcp-server. To run it over stdio from an MCP client, use "command": "docker" with "args": ["run", "-i", "--rm", "-e", "MCP_TRANSPORT_TYPE=stdio", "ghcr.io/cyanheads/worldbank-mcp-server:latest"]. To build locally, docker build -t worldbank-mcp-server .; OpenTelemetry peer dependencies are installed by default — build with --build-arg OTEL_ENABLED=false to omit them.
See CLAUDE.md for development guidelines and architectural rules. The short version:
try/catch in tool logicctx.log for logging, ctx.state for storagecreateApp() arraysIssues are welcome. Run checks and tests before submitting:
This project is licensed under the Apache 2.0 License. See the LICENSE file for details.
Source: README.md at commit 20edda6
List World Bank Topics
List all 21 World Bank thematic topics (Economy & Growth, Health, Education, and others) with descriptions. Use it to browse the indicator space by theme or to find a topic_id for worldbank_search_indicators.
Takes no parameters.
List World Bank Data Sources
List the 70+ World Bank data sources (datasets), such as World Development Indicators, International Debt Statistics, and Doing Business. Returns source IDs and names for use as source_id in worldbank_search_indicators, worldbank_get_data, and worldbank_list_countries, one page at a time.
pageper_pageList World Bank Countries
List countries and regional aggregates with metadata: ISO codes, region, income level, lending type, capital, and coordinates. Filter by region code (e.g. EAS, SSF, NAC), income level (LIC, LMC, UMC, HIC), and lending type (IDX for IDA, IBD for IBRD, IDB for Blend); filters combine by AND. Aggregate entries are excluded by default, leaving individual countries only; set include_aggregates=true to also return region, income group, and world aggregate entities.
pageregionper_pagesource_idincome_levellending_typeinclude_aggregatesGet World Bank Country
Fetch metadata for one country or aggregate entity: ISO codes, region, income level, lending type, capital, and coordinates. Accepts an ISO2 code (US, DE), an ISO3 code (USA, DEU), or a World Bank aggregate code (EAS, HIC, WLD); "all" and lists of codes are rejected. Use worldbank_list_countries to browse valid codes.
country_coderequiredSearch World Bank Indicators
Search the 29,500+ World Bank indicator catalog by keyword, topic, or source, returning indicator IDs and metadata for worldbank_get_data. Provide at least one of query, topic_id, or source_id; a topic and a source together narrow to indicators in both. A keyword query matches every term against indicator ID, name, and full description, in any word order, across the whole catalog or the whole selected topic or source; punctuation is ignored, so the query needs at least one letter or digit. Exact ID or name matches rank first, then names starting with the phrase as whole words, then whole-phrase matches, then matches where every term is a whole word of the ID or name, then other ID/name matches, then description-only matches; within each of those, World Development Indicators series come first, then other live sources, then archives, and a series comes ahead of its own breakdowns by sex, area, or age. Each indicator ID appears once, even where the catalog publishes it under two sources. Each result carries the first 150 characters of its description; worldbank_get_indicator returns the whole of it. Find topic IDs with worldbank_list_topics and source IDs with worldbank_list_sources.
pagequeryper_pagetopic_idsource_idGet World Bank Indicator
Fetch metadata for one World Bank indicator by ID: name, description, unit, source dataset, source organization, and thematic topics. Use worldbank_search_indicators to find the ID when only the concept is known; "all" and lists of IDs are rejected.
indicator_idrequiredGet World Bank Indicator Data
Query World Bank indicator values for one or more countries across a time range — the primary data-access tool; find indicator_id values with worldbank_search_indicators. Observations carry a null value where data is not available for a country×year cell, which is common for sparse series. Set at most one of date_range (a time window), mrv (the latest N periods across the requested countries), or mrnev (each country's own latest N values). Some series, such as Global Economic Monitor's, publish quarterly or monthly values beside annual ones: frequency picks the form mrv and mrnev select from (annual by default), and a date_range's own form (2024, 2024Q1, 2024M01) picks it for a window. lastUpdated gives the serving source's last update date, the data vintage to cite. For "all" countries, page through the results at up to 1000 per page, since the API returns several hundred entries per indicator. Indicators the standard data endpoint does not serve — WDI Database Archives, PEFA, ICP, GDLD, International Debt Statistics: DSSI, Food Prices for Nutrition — are answered from their own dataset instead; the response then carries sourceScoped, naming that dataset and the release, classification, sector, or counterpart area applied (see dimension_value), because those values can be archived or superseded figures rather than current ones.
mrvpagemrnevper_pagecountriesrequiredfrequencysource_iddate_rangeindicator_idrequireddimension_valueGet World Bank Poverty and Inequality Estimates
Query poverty and inequality estimates from the World Bank Poverty and Inequality Platform (PIP) for economies and for PIP's own aggregates — World, World Bank regions, income groups, lending groups, and fragility groups. Returns the poverty headcount ratio, poverty gap, and poverty severity at any poverty line, plus mean and median welfare and population. Use it for inequality and distribution questions too — survey-based economy rows carry the Gini coefficient, mean log deviation, polarization, and the ten decile income/consumption shares, because PIP returns poverty and inequality in the same row. PIP is a separate dataset from the WDI series worldbank_get_data reads: it measures welfare in PPP dollars per person per day, at a PPP vintage ppp_version selects, and computes its aggregates at any poverty line, where worldbank_get_data carries only the published lines. Every row reports how it was produced. Nullable estimateType separately identifies actual, nowcast, or projection estimates, while distributionType preserves the source distribution classification (micro, group, or imputed); missing upstream classifications remain null. On an economy, estimationType "survey" rows carry the full inequality block; "interpolation", "extrapolation", and "CMD estimation" rows are gap-filled estimates for years no survey covers, and their gini, mld, polarization, and decileShares are null — a documented gap in the source data, not an error. Aggregate rows (isAggregate true) are "actual", "nowcast", or "projection", add popInPoverty, and carry no distributional block.
pageyearper_pagecountriesrequiredfill_gapsppp_versionpoverty_linewelfare_typereporting_levelSearch World Bank Projects
Search the World Bank lending portfolio — the individual loans, credits, and grants the Bank finances — by free text, country, region, status, sector names, financing window (IBRD, IDA, Grants, Other), and board approval date. For classified education commitments use sector: ["Education"], rather than a free-text education query. Returns the project ID, name, borrowing country, region, status, board approval and closing dates, the commitment amount in USD with its IBRD, IDA, and grant parts, financing windows, major sectors, and a link to the project page. This is the operations catalogue, not the statistics catalogue: use it for "what is the World Bank funding in Kenya", "which climate adaptation projects are active", or "how much was committed to education in South Asia since 2020". For development statistics and time series, use worldbank_search_indicators and worldbank_get_data instead. Countries take the ISO3 or ISO2 codes the other tools take (BRA or BR), and multi-country operations carry a World Bank regional code such as 3A; WDI aggregates such as SSF or WLD are rejected, since the portfolio lists operations by economy — use region for those. Every filter is an exact match upstream and combines with the others by AND, so a narrow search can legitimately return nothing; when it does, the response says whether the country codes matched anything on their own.
pagequeryregionsectorstatusper_pagecountriesapproved_toapproved_fromfinancial_typeinclude_abstract100% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.
claude mcp add --transport http worldbank-mcp-server https://worldbank.caseyjhand.com/mcpclaude mcp add worldbank-mcp-server -- npx -y @cyanheads/worldbank-mcp-server{ "mcpServers": { "worldbank-mcp-server": { "command": "npx", "args": ["-y", "@cyanheads/worldbank-mcp-server"] } }}git clone https://github.com/cyanheads/worldbank-mcp-server.gitcd worldbank-mcp-server && bun installcp .env.example .env # optional overrides# One-time buildbun run rebuild
# Run the built serverbun run start:http# orbun run start:stdiobun run devcheck # Lint, format, typecheck, and morebun run test # Runs the test suitedocker run --rm -p 3010:3010 ghcr.io/cyanheads/worldbank-mcp-server:latestbun run devcheckbun run test