# Datos Cl [Health: Active]

**Category:** 📊 Data Platforms  
**Repository:** https://github.com/pipeworx-io/mcp-datos-cl  
**GitHub Stars:** 0  
**npm Downloads (last month):** 299  
**Views:** 2  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/datos-cl

## Description
datos.gob.cl CKAN MCP.

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

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

## Documentation

## What Datos Cl does

Datos Cl provides MCP access to the Chilean government’s open-data catalogue at datos.gob.cl. The catalogue runs on CKAN, and the server presents a CKAN-oriented tool surface for finding datasets and examining the organizations, groups, tags, and resources associated with them.

The integration is unofficial and operates through Pipeworx. It is useful when an AI client needs to locate Chilean public datasets, inspect package metadata, identify a dataset’s resources, or browse catalogue structure without building direct CKAN API requests.

## How it works

The hosted MCP endpoint is available at `https://gateway.pipeworx.io/datos-cl/mcp`. An MCP client connects to that URL and receives the Datos Cl tools. The underlying catalogue API is `https://datos.gob.cl/api/3/action`.

Pack-specific connections also expose shared Pipeworx meta-tools, including `ask_pipeworx`, `discover_tools`, `search_within`, and memory-related tools. These additional tools allow a scoped connection to route questions across the wider Pipeworx catalogue. The full gateway is available at `https://gateway.pipeworx.io/mcp` when an application should connect to all packs directly.

For applications that do not use MCP, Pipeworx provides HTTP tool calls under `/v1/tools/`. For example, a client can call `datos_cl_package_list` with JSON parameters over POST. Tool definitions can be inspected with a GET request, and pack discovery is available through the gateway’s search endpoint.

## Setup and configuration

Add the hosted endpoint to an MCP client configuration:

```json
{
  "mcpServers": {
    "datos-cl": {
      "url": "https://gateway.pipeworx.io/datos-cl/mcp"
    }
  }
}
```

The npm package, `@pipeworx/mcp-datos-cl`, runs locally as a stdio process and forwards requests to the same hosted endpoint. A typical configuration starts it with `npx`:

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

Set `PIPEWORX_API_KEY` in the client’s environment to use an account key. Set `PIPEWORX_META_TOOLS=1` with the npm bridge when the shared gateway tools should appear alongside the pack tools. The local bridge requires network access and has no offline mode. When using an HTTPS proxy, Node may require `NODE_USE_ENV_PROXY=1` for its built-in `fetch` implementation to honor proxy settings.

## Tools and capabilities

- List datasets with pagination using `package_list`.
- Search datasets with text, filters, sorting, pagination, and facet fields through `package_search`.
- Inspect a dataset with `package_show` or retrieve an individual resource with `resource_show`.
- Browse organizations and groups, optionally including their datasets.
- List tags, find recently changed packages, and use the gateway’s natural-language routing tools.

## Limitations and notes

Unauthenticated calls are available initially and are capped at 50 calls per day. A free Pipeworx account raises the allowance to 200 calls per day; authenticated requests use a Bearer token and are metered through Pipeworx. The gateway’s shared tools are included on hosted pack connections, so the tool list is larger than the Datos Cl-specific set. The package is an independent integration and is not affiliated with the Chilean government catalogue provider.

_Full upstream README: https://allmcps.com/mcp/datos-cl/readme_