{
  "id": 16492876,
  "title": "Park Graph MCP: let Claude and ChatGPT find, price and book parking",
  "url": "https://urgent.news/2026/10/25/park-graph-mcp-let-claude-and-chatgpt-find-price-and-book-parking",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-25T15:51:12.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/member_86ddbcc1/park-graph-mcp-let-claude-and-chatgpt-find-price-and-book-parking-7i2"
  },
  "original_language": "en",
  "account": "Parking lots often remain hidden from AI assistants. When asked to locate parking near Coors Field, Claude or ChatGPT is forced to make educated guesses based on web pages. To address this issue, Park Graph was created. Owners of parking lots simply need to print one QR sign, which allows drivers to scan and pay within 30 seconds. Every lot on Park Graph is exposed to AI agents through a Model Context Protocol (MCP) server. The MCP endpoint, https://parkgraph.com/api/mcp, is accessible remotely and can be streamed via HTTP.\n\nThe driver experience is seamless, as it requires no API key and is listed in the official MCP Registry as com.parkgraph/mcp, along with its npm package, parkgraph-mcp.\n\nPark Graph offers several tools to facilitate the parking search process. The \"search_lots\" tool provides available parking lots near a specified latitude/longitude, ranked by proximity, along with the number of remaining spaces and current rate. The \"get_lot\" tool offers detailed information about a specific lot, including address, hours, features, pricing, and live space availability.\n\nFor pricing, the \"quote\" tool calculates the cost for a given number of hours at the current rate, applying any daily maximum. The \"reserve\" tool holds a parking space for a few minutes while the driver confirms. The \"get_session\" tool provides status, times, amount, and extensions for an active parking session. Finally, the \"extend_session\" tool allows the driver to add more time to an existing hourly session.\n\nTo integrate Park Graph with AI assistants, simply configure the MCP server and client settings. For instance, the following JSON snippet demonstrates how to connect Claude Desktop, Cursor, or any MCP client that supports remote servers:\n\n{\n\"mcpServers\": {\n\"parkgraph\": {\n\"url\": \"https://parkgraph.com/api/mcp\"\n}\n}\n}\n\nOptional Park Graph API key, either with PARKGRAPH_API_KEY or Authorization: Bearer in the remote endpoint, enables live availability and pricing, and activates session tools. If no parking lots are available, the assistant will respond accordingly.\n\nThe Park Graph solution was designed to eliminate the need for expensive gates and pay stations, which can cost $15,000 to $30,000 per lot. A QR sign, printed at a minimal cost, is sufficient for enabling a convenient booking flow. Owners set the pricing, and the AI assistant handles the rest.\n\nFor more information, visit the Park Graph developers' page at https://parkgraph.com/developers/mcp or their main site at https://parkgraph.com. A free starter plan is available, offering 10% per transaction, up to a maximum of 3 lots. For inquiries, contact support@parkgraph.com.",
  "summary": "Most parking lots are invisible to AI assistants. Ask Claude or ChatGPT to \"find me parking near Coors Field tonight\" and it can only guess from web pages. We built Park Graph to fix that. Lot owners print one QR sign. Drivers scan and pay in 30 seconds. And every lot on Park Graph is exposed to AI agents through a hosted Model Context Protocol (MCP) server. The endpoint…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}