Fra Rail Safety

@pipeworx/fra-rail-safety

US railroad accidents, derailments and highway-rail grade crossing collisions reported to the Federal Railroad Administration, 1975 to the present: railroad, FRA cause code, location, casualties and damage per report, plus grouped counts.

Part of Pipeworx — an MCP gateway connecting AI agents to 1866+ live data sources. This is an independent, unofficial integration — not affiliated with, endorsed by, or published by the upstream provider.

Tools

• rail_accidents(year?, state?, railroad?, cause_code?, accident_type?, min_damage_usd?, limit?) — individual rail equipment accident reports (derailments, collisions, obstructions, fires), newest first, with the reporting railroad, accident type, FRA primary and contributing cause code and text, state/county, train speed, hazmat cars released, people evacuated, killed, injured, damage cost and a link to FRA's own report page.
• crossing_incidents(state?, county?, city?, year?, railroad?, crossing_id?, highway_user?, fatal_only?, limit?) — collisions between trains and road users at railroad crossings, newest first, with the USDOT crossing id, road name, road user type and action, vehicle and train speed, warning devices, weather, killed and injured.
• rail_accident_summary(group_by, dataset?, year?, from_year?, state?, railroad?, limit?) — counts and casualty/damage totals grouped by railroad, cause, accident type, state, year or road user, for either dataset.

Auth

Keyless.

Data sources

• Federal Railroad Administration, Office of Safety Analysis — Rail Equipment Accident/Incident Data (Form 54).
• Federal Railroad Administration, Office of Safety Analysis — Highway-Rail Grade Crossing Incident Data (Form 57).

Both are published by the US Department of Transportation and queried live on each call. US federal government work, public domain.

Things worth knowing:

• One row per railroad report, not per accident. An accident involving two railroads (joint operations, trackage rights) is filed once by each. FRA's unique-accident count keys on the incident key; rail_accident_summary reports both accidents (distinct) and railroad_reports.
• Railroad filter: 1-4 character input is matched against the reporting mark exactly (NS, UP, BNSF, CSX, ATK); longer input is a name substring (Norfolk Southern). A substring match on a reporting mark is wrong — NS occurs inside "CSX TRANSPORTATION".
• State fields differ between the datasets: accidents carry a two-letter code, crossings only the upper-case full name. The pack accepts either form and translates.
• Lag: railroads file monthly and FRA publishes a few months behind, so the current year is partial.
• Cause codes are FRA's: the first letter is the family (T track, S signal, H human factors, E equipment, M miscellaneous); cause_code matches a prefix.

Quick Start

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

{
  "mcpServers": {
    "fra-rail-safety": {
      "url": "https://gateway.pipeworx.io/fra-rail-safety/mcp"
    }
  }
}

What this endpoint actually serves

tools/list at https://gateway.pipeworx.io/fra-rail-safety/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 1866+ 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/rail_accidents \
  -H 'Content-Type: application/json' \
  -d '{"year":2023,"state":"OH","railroad":"NS","limit":5}'

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

Standalone (no gateway account)

This package also runs as a local stdio MCP server — no Pipeworx account, no gateway round-trip:

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

Or run it directly to confirm it starts:

npx -y @pipeworx/mcp-fra-rail-safety

It speaks MCP over stdin/stdout and answers initialize/tools/list/tools/call for only this pack's tools — none of the shared meta-tools the gateway connection above adds. Same source, same tools, no ask_pipeworx routing.

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 Fra Rail Safety data" })

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

More

• Docs and guides
• pipeworx.io

License

MIT