@pipeworx/idb

Inter-American Development Bank (IDB) project procurement — live bidding notices, contract awards, and a 5-multilateral-development-bank cross-debarment check, all sourced from IDB's keyless CKAN open-data portal.

Part of Pipeworx — an MCP gateway connecting AI agents to 1865+ live data sources. This is an independent, unofficial integration — not affiliated with, endorsed by, or published by the upstream provider.

Tools

• idb_search_notices(...) — search live IDB procurement notices (bids, expressions of interest, general/specific notices, award notifications) across 26 LAC countries; defaults to open (deadline not passed) notices.

• idb_get_notice(notice_id) — full record for one notice.

• idb_search_awards(...) — search ~159k awarded IDB project contracts by country, firm, sector, or year.

• idb_check_mdb_debarment(name, ...) — check a name against IDB's own debarment list plus the cross-debarment lists of the World Bank Group, Asian Development Bank, African Development Bank, and EBRD (one shared feed answering "is this vendor debarred" across five banks).

Related MCP server: mcp-open-contracting

Auth

Keyless.

Data sources

• https://data.iadb.org/dataset/project-procurement-bidding-notices-and-notification-of-contract-awards — bidding notices + award notifications (~37,000 rows).

• https://data.iadb.org/dataset/idb-project-procurement-contract-awards-data — award-level contract detail (~159,000 rows; the dataset page offers a ~70MB CSV, but the same data is exposed via CKAN's Datastore, so every tool here queries it live with server-side SQL rather than downloading the file).

• https://data.iadb.org/dataset/dataset-of-sanctioned-firms-and-individuals — IDB debarment + WBG/ADB/AfDB/EBRD cross-debarment (~1,300 rows).

All three are queried through data.iadb.org/api/3/action/datastore_search_sql (CKAN Datastore, keyless, no auth). Notes for anyone touching this pack:

• Every resource here is CKAN Datastore-active — package_show on each dataset reports datastore_active: true, which is what makes live SQL querying possible instead of downloading and caching the 70MB awards file ourselves. Check datastore_active before assuming a new IDB dataset needs downloading.

• The plain /files/download/<id> URLs sit behind an AWS WAF bot challenge (HTTP 202, x-amzn-waf-action: challenge, on a bare request) — irrelevant here since nothing calls them, but don't reach for them if extending this pack; the Datastore API is unaffected.

• The upstream's own full-text search (q= on datastore_search) is accent-SENSITIVE. q=Gonzalez (unaccented) returns zero rows against a debarment table full of "González" (with the accent) — verified live. idb_check_mdb_debarment never uses q=; it fetches the whole (small, ~1,300-row) debarment table and matches client-side with NFD-normalized, diacritic-stripped, token-substring comparison.

• type in the notices table has inconsistent trailing whitespace — SELECT DISTINCT "type" returns "AWARD", "AWARD ", and "AWARD " as three separate values. Always TRIM() before comparing, and match with a prefix ILIKE, never =.

• datastore_search_sql has no bind-parameter support — every string argument is quote-escaped (' → '') before interpolation. CKAN's SQL endpoint itself only accepts SELECT statements against a read replica.

• Many idb-project-procurement-contract-awards-data rows record awarded_firm_name as "Not Available" — this is by design for individual-consultant selections (contract_type: "Individual Consultants"), not a data gap.

Quick Start

Add to your MCP client (Claude Desktop, Cursor, Windsurf, etc.):

{
  "mcpServers": {
    "idb": {
      "url": "https://gateway.pipeworx.io/idb/mcp"
    }
  }
}

What this endpoint actually serves

tools/list at https://gateway.pipeworx.io/idb/mcp returns the tools in the table above plus the shared Pipeworx meta-tools — ask_pipeworx, discover_tools, search_within, remember/recall and the rest of the gateway-wide set. So the tool count you see is larger than this table: a single-pack endpoint currently lists roughly 30 shared tools alongside the pack's own. The connection's initialize response states its exact scope, and is the authoritative answer for a given day.

This is deliberate, not multiplexing by accident. The meta-tools are what let a scoped connection answer a question this pack does not cover — via ask_pipeworx, which routes across the whole catalog — without you adding a second MCP server. There is currently no way to mount a pack endpoint without them; if the extra schemas cost you more context than the routing is worth, connect to the full gateway once rather than to several pack endpoints.

Or connect to the full Pipeworx gateway to get every pack's tools listed directly, instead of just this one's:

{
  "mcpServers": {
    "pipeworx": {
      "url": "https://gateway.pipeworx.io/mcp"
    }
  }
}

Both URLs reach the same gateway and the same 1865+ data sources. The only difference is which pack's tools are listed directly; ask_pipeworx reaches all of them from either one.

No MCP client? Call it over HTTP

curl -X POST https://gateway.pipeworx.io/v1/tools/idb_search_notices \
  -H 'Content-Type: application/json' \
  -d '{"country":"Peru","open_only":true,"limit":5}'

No account needed for the first calls. Inspect any tool: GET https://gateway.pipeworx.io/v1/tools/idb_search_notices. Find one: POST https://gateway.pipeworx.io/v1/tools/search_packs with {"query":"..."}.

Run it from npm

@pipeworx/mcp-idb is a local stdio bridge to the Pipeworx gateway: your MCP client starts it with npx, and it forwards every call to the same https://gateway.pipeworx.io/idb/mcp endpoint shown above. It needs network access; there is no offline mode.

{
  "mcpServers": {
    "idb": {
      "command": "npx",
      "args": ["-y", "@pipeworx/mcp-idb"]
    }
  }
}

Without a key it calls anonymously, capped at 50 calls a day. A free account raises that to 200 a day: get a key at https://pipeworx.io/signup and add "env": { "PIPEWORX_API_KEY": "<your key>" } to the server entry. The key is sent as a Bearer token, the same way a hosted connection sends it, so calls are metered and billed exactly as they are there.

It lists only this pack's tools by default. Set PIPEWORX_META_TOOLS=1 to also list ask_pipeworx and the other gateway-wide tools.

Behind an HTTPS proxy: Node's built-in fetch does not read HTTPS_PROXY by default. Newer Node releases honour it when NODE_USE_ENV_PROXY=1 is set; check your Node version's documentation.

Using with ask_pipeworx

Instead of calling tools directly, you can ask questions in plain English — this works on the pack endpoint above as well as on the full gateway:

ask_pipeworx({ question: "your question about Idb data" })

The gateway picks the right tool and fills the arguments automatically.

More

• Docs and guides

• pipeworx.io

License

MIT