Privacy Policy — NYCfoodie

Effective date: 2026-10-04
Contact: tireniajilore1@gmail.com

NYCfoodie is a free, public MCP (Model Context Protocol) server that gives AI agents structured editorial restaurant recommendations for New York City. There are no user accounts, no sign-up, no cookies, and we do not ask for any personal information.

What we collect

1. Anonymous usage analytics — stored in a database table, automatically deleted after 180 days:

• timestamp of the call
• tool name (e.g. search_restaurants)
• coarse city parameter passed to the tool
• request latency and whether it succeeded
• a 16-character truncated SHA-256 hash of your IP address plus user-agent string. This lets us count distinct clients without storing IPs — the hash cannot be reversed into an IP address or device.
• the connecting client's self-reported software name and version (from the MCP handshake), and a reduced user-agent string with platform details stripped (e.g. Mozilla/5.0 without OS/version detail).

This analytics data contains no query text and no raw IP addresses.

2. Operational server logs — one line per tool call, kept on the server:

• tool name, the arguments sent to the tool, duration, success or error, and result size.

These logs exist for debugging, reliability monitoring and abuse prevention. They are not shared, not sold, and not exposed through any public endpoint. Unlike the analytics above, they do include the parameters your agent sent (e.g. a cuisine or neighbourhood search) — but they never include anything beyond what your agent itself transmitted to the server.

3. Feedback you choose to send — if your agent calls submit_feedback:

• the tool name, a 1–5 rating, and the comment text your agent wrote.

Feedback is voluntary, stored indefinitely so we can improve the service, and never linked to an identity — we have no accounts to link it to.

What we do not do

• No raw IP addresses are stored.
• No personal data is requested or required.
• Data is not sold, rented, or shared with third parties.
• The admin analytics endpoints are protected and visible only to the operator.

Data provenance note (for transparency)

NYCfoodie's restaurant data is derived from The Infatuation's published restaurant guides, transformed into structured data (ratings, tags, booking intel). Data licensing is currently unresolved — see the project repository.

Hosting

The service runs on Railway (United States). Standard server infrastructure (Railway) may process requests in the course of hosting.

Changes

If this policy changes materially, the updated version will be published at this URL with a new effective date.