Server data from the Official MCP Registry
Police Shootings MCP — The Washington Post, Fatal Force database
About
Police Shootings MCP — The Washington Post, Fatal Force database
Remote endpoints: streamable-http: https://gateway.pipeworx.io/police-shootings/mcp
Security Report
This server is a read-only proxy for The Washington Post Fatal Force CSVs fetched from raw.githubusercontent.com, with no credentials or shell, file, or env access in the reviewed code. The main concerns are that the police_shootings tool returns names of people killed as published, which is personal data even though it is public, and that the src/index.ts file was truncated, so the remaining tool handlers and bin/cli.js were not reviewed. Supply chain analysis found 1 known vulnerability in dependencies (0 critical, 1 high severity).
3 files analyzed · 6 issues found
Security scores are indicators to help you make informed decisions, not guarantees. Always review permissions before connecting any MCP server.
Permissions Required
This plugin requests these system permissions. Most are normal for its category.
How to Install & Connect
Available as Local & Remote
This plugin can run on your machine or connect to a hosted endpoint. during install.
Documentation
View on GitHubFrom the project's GitHub README.
@pipeworx/police-shootings
U.S. fatal police shootings since 2015 — aggregate counts, individual incident records (including the name of the person killed, as published), and per-agency breakdowns, proxied live from The Washington Post's Fatal Force database.
Part of Pipeworx — an MCP gateway connecting AI agents to 1784+ live data sources. This is an independent, unofficial integration — not affiliated with, endorsed by, or published by the upstream provider.
Tools
police_shootings_summary(group_by?, state?, year?)— aggregate counts (by state, year, race, armed_with, threat_type, gender, mental_illness, or body_camera). Never returns names.police_shootings(state?, city?, year?, armed_with?, race?, limit?, offset?)— individual incident records, including the name of the person killed. Small by default (limit 25, max 200); a truncated response carriestruncated:true,total_matching, and a narrowing hint.agency_shootings(agency, state?)— one law-enforcement agency's fatal-shooting count and breakdown by year/race/armed_with. Matches on a name fragment; returns a small list of candidates when more than one agency matches, rather than guessing.
Auth
Keyless. The upstream is a public CSV with no registration, key, or login.
Data sources
- The Washington Post, Fatal Force database — fatal shootings by on-duty
police officers in the United States since Jan. 1, 2015, plus the roster
of law-enforcement agencies involved. Published under
Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0);
every tool response carries
licenseandattributionfields so the terms travel with the data. This pack is a LIVE PROXY — it fetches the Post's published files fresh on every cache miss (6h edge cache, same as any normal client would cache) and never commits the dataset into this repo (CLAUDE.md "mirror-needs-grant-proxy-does-not": a per-request proxy call is not redistribution; bulk-copying the dataset would be, and the licence forbids that). Human-facing page: https://www.washingtonpost.com/graphics/investigations/police-shootings-database/. Routes live only insrc/index.ts.
Notes
- The Post's own README states the data is "updated regularly," but the
published file this pack reads currently ends 2024-12-31 (verified live
2026-10-13) — there is no newer data to proxy. Every response's
data_as_of/dataset_coverage.max_datereflects whatever the Post has actually published at call time, not a fixed claim. - The Post tracks only on-duty police shootings paralleling the Michael Brown case — not deaths in custody, off-duty shootings, or non-shooting deaths. It is not comprehensive beyond that scope (the Post itself notes undercounting is possible before reporting catches up).
raceandarmed_withcan carry more than one value per incident, semicolon-separated in the source (e.g."W;B","gun;knife"). Theraceandarmed_withfilters inpolice_shootings/police_shootings_summarymatch if the requested value is ANY one of the semicolon-separated tokens.agency_idson an incident can list more than one agency (semicolon-separated) when multiple departments were involved;agency_shootingsjoins on that list.- Names, ages, and other person-level fields are served exactly as the Post publishes them — nothing is added, inferred, or joined against any other people-data source.
Quick Start
Add to your MCP client (Claude Desktop, Cursor, Windsurf, etc.):
{
"mcpServers": {
"police-shootings": {
"url": "https://gateway.pipeworx.io/police-shootings/mcp"
}
}
}
What this endpoint actually serves
tools/list at https://gateway.pipeworx.io/police-shootings/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 1784+ 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/police_shootings_summary \
-H 'Content-Type: application/json' \
-d '{"group_by":"year"}'
No account needed for the first calls. Inspect any tool: GET https://gateway.pipeworx.io/v1/tools/police_shootings_summary. 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": {
"police-shootings": {
"command": "npx",
"args": ["-y", "@pipeworx/mcp-police-shootings"]
}
}
}
Or run it directly to confirm it starts:
npx -y @pipeworx/mcp-police-shootings
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 Police Shootings data" })
The gateway picks the right tool and fills the arguments automatically.
More
License
MIT
Reviews
No reviews yet
Be the first to review this server!
More Developer Tools MCP Servers
Fetch
Freeby Modelcontextprotocol · Developer Tools
Web content fetching and conversion for efficient LLM usage
Git
Freeby Modelcontextprotocol · Developer Tools
Read, search, and manipulate Git repositories programmatically
Worldmonitor
Freeby Koala73 · Developer Tools
Live markets, conflicts, country risk, chokepoints, energy, and China decision signals. 93 tools.
Toleno
Freeby Toleno · Developer Tools
Toleno Network MCP Server — Manage your Toleno mining account with Claude AI using natural language.
MCP Marketplace
Freeby mcp-marketplace · Developer Tools
Search and install MCP servers from inside your AI client.
mcp-creator-python
Freeby mcp-marketplace · Developer Tools
Create, build, and publish Python MCP servers to PyPI — conversationally.
