LPG Connector MCP Server
Enables Claude to query live LPG price and stock data, find nearest stations, view market trends, and submit crowdsourced price reports via MCP tools.
README
LPG Connector — MCP Server
A remote MCP server that exposes live LPG price/stock data to Claude. Backed by in-memory mock data for now; built to swap to MongoDB with no changes to the tool logic.
Tools
| Tool | What it does |
|---|---|
get_lpg_price(region?) |
Latest price/kg + stock per station in a region (omit region for all) |
find_nearest_station(lat, lng, limit?) |
Nearest stations to a coordinate, with distance + latest price |
get_market_trends(region, days?) |
Daily average price trend over the last N days |
report_price(stationId, pricePerKg, stock, reporter?) |
Submit a crowdsourced price/stock report |
stock is one of in_stock | low | out_of_stock.
Run locally
npm install
npm run dev # tsx watch, http://localhost:3000/mcp
Health check: GET http://localhost:3000/health
Quick smoke test (list tools):
curl -s -X POST http://localhost:3000/mcp \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/list","params":{}}'
Build for production
npm run build # -> dist/
npm start # node dist/index.js
Project layout
src/
index.ts Express + streamable HTTP transport (the /mcp endpoint)
server.ts MCP server + the 4 tool definitions
store/
types.ts Domain types (Station, PriceReport, ...)
DataStore.ts Interface every store implements
InMemoryStore.ts JSON-seeded implementation (current)
data/seed.json Mock stations + price history
Swapping to MongoDB later
The tools only ever talk to the DataStore interface, so the migration is
localized:
npm install mongodb- Create
src/store/MongoStore.tsimplementingDataStore(same method signatures asInMemoryStore). MapstationsandpriceReportsto two collections. - In
src/index.ts, change one line:// const store = new InMemoryStore(); const store = new MongoStore(process.env.MONGODB_URI!);
Nothing in server.ts changes. Seed the collections from data/seed.json once
for a matching demo dataset.
Deploy (for the Claude custom connector)
Claude's cloud connects to this server over the public internet — a localhost URL won't work for claude.ai. This repo is set up for Render.
Deploy to Render
Option A — Blueprint (uses render.yaml, one click):
- Push this project to a GitHub repo.
- In Render: New → Blueprint, pick the repo. It reads
render.yaml(buildnpm install && npm run build, startnpm start, health check/health). - Deploy. You'll get a URL like
https://cylindr.onrender.com.
Option B — Manual web service:
- New → Web Service, connect the repo, set Root Directory to
zserver. - Build command:
npm install && npm run build - Start command:
npm start - Health check path:
/health
After it's live
- Confirm health: open
https://<your-app>.onrender.com/health→{"status":"ok"} - Add to Claude: Settings → Connectors → Add custom connector, paste
https://<your-app>.onrender.com/mcp, then enable it in a chat via the + menu.
Free-tier note: Render's free web services sleep after ~15 min idle and take ~30–60s to wake on the next request. Before your live demo, hit
/healthonce to wake it so the first judge query isn't slow. Also: in-memory data is per-process, so a restart (or wake from sleep) resets to the seed and anyreport_pricesubmissions are lost. The MongoDB swap removes both concerns.
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.