# Transport Feeds [Health: Active]

**Category:** 🚆 Travel & Transportation  
**Repository:** https://github.com/pipeworx-io/mcp-transport-feeds  
**GitHub Stars:** 0  
**npm Downloads (last month):** 461  
**Views:** 1  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/transport-feeds

## Description
Transport Feeds MCP.

## Tools
Capabilities this server exposes over MCP:

- **list_feeds** — List the curated transport & mobility feeds (id, title, category, source). Optionally filter by category (automotive, transport) or keyword. Pass an id to read_feed.
- **read_feed** — Read a curated transport & mobility feed by its id (from list_feeds). Returns normalized items (title, link, published, summary). Optionally filter items by keyword.
- **fetch_feed** — Fetch and normalize any RSS / Atom / RDF feed by URL. CF-robust: fetches directly and falls back to a proxy if the source blocks the gateway. Use list_feeds first for curated sources.

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

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

## Documentation

## What Transport Feeds does

The Transport Feeds MCP server gives AI agents access to curated transport and mobility feeds. It also accepts feed URLs outside the curated collection, allowing an agent to work with a specific RSS, Atom, or RDF source when needed.

Curated sources include an ID, title, category, and source URL. Categories can be narrowed to `automotive` or `transport`, and keyword filtering is available during discovery. Retrieved entries are normalized to include a title, link, publication date, and summary.

This fits workflows that monitor transportation coverage, look up mobility updates, or collect feed items for further analysis. It does not present a general transport database or a live vehicle-tracking interface; the documented capabilities are feed discovery, retrieval, normalization, and filtering.

## How it works

The Transport Feeds MCP server exposes three tools:

- `list_feeds` discovers the available curated feeds. Callers can filter by category or keyword. The returned feed ID is used with `read_feed`.
- `read_feed` reads one curated feed by ID and returns normalized items. Item results can be filtered by keyword.
- `fetch_feed` retrieves and normalizes any RSS, Atom, or RDF feed supplied by URL. It first attempts a direct fetch and can use a proxy fallback when the source blocks the gateway.

On Pipeworx, the scoped endpoint also lists shared gateway meta-tools, so its reported tool count is larger than these three pack-specific tools. The gateway's `initialize` response is the authority for the tools available on a particular connection. The `ask_pipeworx` meta-tool can route questions across the wider Pipeworx catalog.

## Setup and configuration

Use the remote MCP endpoint at `https://gateway.pipeworx.io/transport-feeds/mcp` in an MCP client. The README specifically shows configurations for Claude Desktop, Cursor, and Windsurf. A full Pipeworx connection is also available at `https://gateway.pipeworx.io/mcp`; both endpoints use the same gateway and data sources, while the scoped endpoint lists this pack directly.

For a local connection, run the package with `npx -y @pipeworx/mcp-transport-feeds` and configure the command as an MCP stdio server. The local process handles `initialize`, `tools/list`, and `tools/call` for only the pack's own tools and does not require a Pipeworx account or gateway round trip.

The tools can also be called over HTTP. For example, the gateway exposes `POST https://gateway.pipeworx.io/v1/tools/transport_feeds_list_feeds` with a JSON body such as `{"category":"automotive"}`. The documented examples state that initial calls do not require an account.

## Tools and capabilities

Transport Feeds MCP server capabilities include:

- Discovering curated transport and mobility feeds.
- Filtering feed discovery by `automotive`, `transport`, or keyword.
- Reading curated feeds by their IDs.
- Filtering returned feed items by keyword.
- Fetching arbitrary RSS, Atom, and RDF URLs.
- Returning normalized item fields for consistent downstream handling.
- Falling back to a proxy when a source blocks the gateway during fetching.

## Limitations and notes

`read_feed` expects a feed ID obtained from `list_feeds`; use `fetch_feed` for an arbitrary URL. The remote pack endpoint includes shared Pipeworx meta-tools in addition to the pack tools, while the standalone package exposes only the three documented feed tools. Feed contents depend on the selected sources and their available entries.

_Full upstream README: https://allmcps.com/mcp/transport-feeds/readme_