got-cosy-mcp
MCP server providing cosiness scores for about 6,300 independent hotels, with tools to find hotels by city/country and retrieve detailed feeling data.
README
got-cosy-mcp
An MCP server for Got Cosy, the Feel Layer for hotel data: cosiness scores for about 6,300 independent hotels, computed from guest reviews and photos, with verified booking links. Free and read-only.
This package is a thin stdio proxy. Each tool call is forwarded over standard MCP JSON-RPC to Got Cosy's hosted Streamable HTTP endpoint at https://gotcosy.com/api/mcp and the result is returned unchanged. There are no local secrets, no API key and no local data: this server only needs outbound HTTPS.
If your client speaks Streamable HTTP directly, you can skip this package entirely and point it at https://gotcosy.com/api/mcp (manifest: https://gotcosy.com/mcp.json). Use this stdio proxy for clients that only support stdio-based MCP servers.
Install and run
Using npx (no install step):
npx got-cosy-mcp
Using Docker:
docker build -t got-cosy-mcp .
docker run -i --rm got-cosy-mcp
Add to an MCP client config (example for a client that reads a JSON config of servers):
{
"mcpServers": {
"got-cosy": {
"command": "npx",
"args": ["-y", "got-cosy-mcp"]
}
}
}
Tools
find_cosy_hotels
Find live, cosy-scored hotels. Only hotels that clear the public floor (score 5.0+) are returned. All arguments are optional.
| Argument | Type | Description |
|---|---|---|
city |
string | City name, e.g. "Paris" |
country |
string | Country name, e.g. "France" |
min_score |
number | Minimum cosy score 0-10 (the public floor of 5.0 always applies) |
limit |
number | Max results, default 20, max 100 |
Example call:
{
"name": "find_cosy_hotels",
"arguments": { "city": "Paris", "limit": 5 }
}
Example result (truncated):
{
"hotels": [
{
"slug": "les-dames-du-pantheon",
"name": "Les Dames du Panthéon",
"city": "Paris",
"country": "France",
"cosy_score": 6.7,
"url": "https://gotcosy.com/en/hotels/les-dames-du-pantheon",
"verified_booking": "stay22",
"website": "https://www.hotellesdamesdupantheon.com"
}
],
"total": 81,
"limit": 5,
"offset": 0
}
get_hotel_feeling
Get one hotel's cosy score, evidence signals and description by its Got Cosy slug (as returned by find_cosy_hotels, or from a gotcosy.com/en/hotels/<slug> URL).
| Argument | Type | Required | Description |
|---|---|---|---|
slug |
string | yes | The hotel's Got Cosy slug |
Hotels below the public floor return {"below_bar": true} with no score exposed.
Data and scoring
Got Cosy scores independent hotels 0-10 for warmth, intimacy and character, drawn from guest reviews and photos rather than star ratings or price. Only hotels that clear a public floor of 5.0 are exposed through these tools. See gotcosy.com for the full index and gotcosy.com/llms-full.txt for machine-readable documentation.
Links
- Hosted endpoint:
https://gotcosy.com/api/mcp(Streamable HTTP, stateless, no auth) - Manifest:
https://gotcosy.com/mcp.json - Website: gotcosy.com
- Source repository: github.com/perwinroth/cosyhotels (official registry entry:
io.github.perwinroth/got-cosy-graph) - Contact: per@gotcosy.com
Licence
MIT, see LICENSE.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.