# Opendosm My [Health: Active]

**Category:** 🧮 Data Science Tools  
**Repository:** https://github.com/pipeworx-io/mcp-opendosm-my  
**GitHub Stars:** 0  
**npm Downloads (last month):** 414  
**Views:** 1  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/opendosm-my

## Description
data.gov.my / OpenDOSM (Department of Statistics Malaysia) — official Malaysian open-data API.

## Tools
Capabilities this server exposes over MCP:

- **opendosm_get_dataset** — Fetch records from an official Malaysian statistics dataset (data.gov.my / OpenDOSM, Dept of Statistics Malaysia). Keyless, authoritative. The dataset `id` is REQUIRED — there is no listing endpoint, so use opendosm_list_datasets or opendosm_dataset_meta to discover ids and their columns. Verified…
- **opendosm_dataset_meta** — Metadata for one Malaysian dataset (data.gov.my): catalogue_id, data_as_of, last_updated, next_update, data_source, update_frequency, plus a sample row showing the available columns. Use this to inspect a series before pulling it, or to check freshness. Requires a dataset id.
- **opendosm_list_datasets** — List the curated, verified high-value datasets available from data.gov.my / OpenDOSM (Dept of Statistics Malaysia). data.gov.my exposes no machine listing/search endpoint, so this returns a hand-verified set covering inflation, GDP, labour, prices, income, population and demography, with each datas…

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "opendosm-my": {
    "command": "npx",
    "args": ["-y","@pipeworx/mcp-opendosm-my"]
  }
}
```

## Documentation

## What Opendosm My MCP server does

Opendosm My MCP server gives MCP-compatible AI clients access to selected datasets from data.gov.my/OpenDOSM, the open-data service of Malaysia’s Department of Statistics. The package focuses on verified, high-value statistical series rather than exposing a general-purpose search interface.

The available catalogue covers subjects including inflation, GDP, labour, prices, income, population and demography. Data retrieval is keyless, so the server does not require an API token for the documented access paths.

## How it works

Opendosm My MCP server exposes three pack-specific tools. `opendosm_list_datasets` returns a curated set of dataset IDs, topics, descriptions and value columns. It can optionally filter the results by topic. Because data.gov.my does not provide a machine listing or search endpoint for this use case, this curated list is the starting point for discovering supported series.

`opendosm_dataset_meta` accepts a dataset ID and returns catalogue and update information, including the data-as-of date, last update, next update, source and update frequency. It also supplies a sample row so an agent can inspect the available columns before requesting records.

`opendosm_get_dataset` fetches records for a required dataset ID. Retrieval supports request options such as a record limit and sort order, as shown by the documented HTTP example. An ID must be supplied; the retrieval tool is not a substitute for dataset discovery.

## Setup and configuration

The Pipeworx gateway exposes the server at `https://gateway.pipeworx.io/opendosm-my/mcp`. Add that URL to an MCP client that supports the documented remote connection format. The README specifically shows Claude Desktop, Cursor and Windsurf configurations.

For a local connection, run `npx -y @pipeworx/mcp-opendosm-my` and configure the client to launch the command. The local process communicates over standard input and output and exposes only the three Opendosm My MCP server tools, without the gateway’s shared meta-tools. No Pipeworx account is required for this standalone mode.

The gateway also offers direct HTTP tool calls under `https://gateway.pipeworx.io/v1/tools/`, including a POST endpoint for `opendosm_get_dataset`. The gateway connection adds shared Pipeworx tools such as discovery and routing; the standalone package does not.

## Tools and capabilities

- Discover curated Malaysian datasets and optionally filter by topic.
- Inspect dataset freshness, provenance, update timing and sample columns.
- Fetch records from a selected dataset by ID.
- Query official OpenDOSM data without supplying an API key.
- Use the tools through MCP, the Pipeworx gateway, or direct documented HTTP calls.

## Limitations and notes

The dataset ID is mandatory when fetching records. There is no machine listing endpoint in the underlying data service, so use the curated list or metadata tool to identify supported IDs and columns first.

The gateway’s pack endpoint lists the pack tools along with shared Pipeworx meta-tools, so its tool count is larger than the three tools documented here. The exact gateway scope can change; the connection’s `initialize` response is authoritative. The local stdio package is limited to this pack and does not provide cross-catalogue routing through `ask_pipeworx`.

_Full upstream README: https://allmcps.com/mcp/opendosm-my/readme_