SafeStreets
RemoteAbout SafeStreets
SafeStreets is an MCP server for Claude, ChatGPT, and other assistants to examine walkability and pedestrian safety across US addresses, cities, and neighborhoods. Its read-only tools are paired with skills for evaluating an address, narrowing places to live, and gathering street-safety evidence.
Key features
- 01Score address walkability and safety
- 02Compare places and score components
- 03Find city and neighborhood scores
- 04Filter neighborhoods by price and transit
- 05Identify roads with pedestrian deaths
- 06Search citable research articles
Use cases
- 01Assess whether a US address is walkable
- 02Narrow relocation options by budget and transit
- 03Gather evidence for a street-safety proposal
MCP support
The SafeStreets MCP server exposes read-only tools for address scores, neighborhood searches, dangerous corridors, and research articles.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://safestreets.streetsandcommons.com/mcpclaude mcp add --transport http poseidon-t-safestreets https://safestreets.streetsandcommons.com/mcp{"mcpServers":{"poseidon-t-safestreets":{"type":"http","url":"https://safestreets.streetsandcommons.com/mcp"}}}Open in Cursor{"servers":{"poseidon-t-safestreets":{"type":"http","url":"https://safestreets.streetsandcommons.com/mcp"}}}Open in VS Code{"mcpServers":{"poseidon-t-safestreets":{"type":"http","url":"https://safestreets.streetsandcommons.com/mcp"}}}{"mcpServers":{"poseidon-t-safestreets":{"serverUrl":"https://safestreets.streetsandcommons.com/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
Details
From the project READMESafeStreets plugin
Walkability and pedestrian safety for any US address, for Claude and ChatGPT. Made by Streets & Commons.
The plugin connects your assistant to the SafeStreets MCP server at https://safestreets.streetsandcommons.com/mcp and adds three skills that tell the assistant how to use it well. It is free and needs no sign-in.
Tools
| Tool | What it answers |
|---|---|
get_walkability_score | How walkable and how safe for walking one address is, 0 to 10, with four components and the metrics behind them |
compare_locations | Which of several places is better on foot, and on which component they differ |
score_many | Scores for a list of addresses |
get_city_walkability | A whole city's score |
find_walkable_neighborhoods | The most walkable neighborhoods in a city |
search_neighborhoods | Neighborhoods across the US filtered by region, home price, walkability and transit |
get_dangerous_corridors | Roads where pedestrians were killed, from NHTSA FARS, with deaths per year |
search_insights | SafeStreets research articles with figures you can cite |
Every tool only reads data.
Skills
- walkability-check: answers "is this address walkable?" with a verdict and the component that holds it back.
- where-to-live-on-foot: narrows a move by region, budget and transit, then compares a shortlist.
- street-safety-evidence: builds a sourced case about a dangerous street for a council letter, story or grant.
Install
Claude Code
/plugin marketplace add Poseidon-t/safestreets-plugin
/plugin install safestreets@safestreets
Claude.ai or Claude Desktop: Settings, Connectors, Add custom connector, then paste https://safestreets.streetsandcommons.com/mcp.
ChatGPT: Settings, Apps and Connectors, Create, then paste the same URL with no authentication.
Cursor or any MCP client
{
"mcpServers": {
"safestreets": { "url": "https://safestreets.streetsandcommons.com/mcp" }
}
}
Limits
Keyless calls share a daily limit on full address scores. A free API key gives you your own 10,000 calls a month: send it as the x-api-key header. Coverage is the United States only.
Data and policies
Scores come from OpenStreetMap, NHTSA crash records, the US Census, EPA and satellite imagery. See the methodology, privacy policy and terms.
Support: [email protected]
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.