Skip to main content
Glama
tw-fda-notices

tw-fda-notices

by pipeworx-io

@pipeworx/tw-fda-notices

Which companies has Taiwan's Food and Drug Administration (食品藥物管理署) named for failed food imports or drug recalls? Two searches over the FDA's published open data. Firms only.

Part of Pipeworx — an MCP gateway connecting AI agents to 2020+ live data sources. This is an independent, unofficial integration — not affiliated with, endorsed by, or published by the upstream provider.

Tools

  • tw_fda_border_search(query?, origin?, date_from?, date_to?, limit?, offset?) -- imported food and food-related products that failed border inspection (邊境查驗不合格), since 2023, newest first. Each notice gives the publication date, origin country, product, importer, foreign manufacturer or exporter, the failure and the measured value against the legal limit, weight and disposition (return or destroy). Also returns total_matching, truncated, searched, skipped_possible_individuals and data_as_of.

  • tw_fda_drug_recall_search(query?, grade?, date_from?, date_to?, limit?, offset?) -- drug recalls (回收藥品) from 2008: product, licence number and holder, lots, reason, recall grade and the FDA document number.

query is matched in Chinese. A search with no hits returns found: false and empty_reason: "no_match" and says how many rows were read; a source that cannot be read raises upstream_down instead.

Related MCP server: FDA API MCP Service

Auth

Keyless.

Data sources

  • Taiwan Food and Drug Administration (食品藥物管理署), open data: non-conforming imported food information (不符合食品資訊).

  • Taiwan Food and Drug Administration (食品藥物管理署), open data: recalled drugs (回收藥品).

Each call reads the publisher's current dataset (about 3 MB for the border notices, 0.7 MB for recalls) and filters it; nothing is stored. The publisher's open-data terms are the government open data licence.

Individuals are not returned

Rows whose company is not clearly a firm are left out and counted in skipped_possible_individuals. Importer street addresses are never returned. The tests prove a person's name and address do not appear and that an all-individual result reports itself rather than looking like no match.

Behaviour worth knowing

  • Only failures are listed; a firm absent from the dataset has not been named there, which is not a clean bill.

  • Importer names that start with a business registration number have the number removed.

  • Matching is on Chinese text as published.

Quick Start

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

{
  "mcpServers": {
    "tw-fda-notices": {
      "url": "https://gateway.pipeworx.io/tw-fda-notices/mcp"
    }
  }
}

What this endpoint actually serves

tools/list at https://gateway.pipeworx.io/tw-fda-notices/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 2020+ 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/tw_fda_border_search \
  -H 'Content-Type: application/json' \
  -d '{"query":"辣椒","limit":3}'

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

Run it from npm

@pipeworx/mcp-tw-fda-notices 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/tw-fda-notices/mcp endpoint shown above. It needs network access; there is no offline mode.

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

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 Tw Fda Notices data" })

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

More

License

MIT

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables querying FDA inspection outcomes, Form 483 observations, compliance actions, and import refusals, plus a combined firm risk profile, from the FDA Data Dashboard.
      60 npm
      MIT
    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables AI agents to search Singapore Food Agency food alerts and recalls — foods recalled or found unsafe for foreign matter, undeclared allergens, contamination, or adulteration — by keyword and date range, returning each notice's date, type, title, product, reason, and link. Data is read live from the Agency's published list on every call, with nothing stored.
      MIT