Connect Your AI Assistant

claude.ai and the Claude desktop apps — all plans

1. Open Settings → Connectors (or claude.ai/customize/connectors).
2. Click Add custom connector.
3. Paste the server URL and click Add. No sign-in is requested — the server is public and read-only.
4. In a chat, enable the connector and ask about inventory, pricing or delivery.

Plus, Pro, Team and Enterprise

1. Settings → Apps → Advanced settings → turn on Developer mode.
2. Settings → Apps → Create. Name it Prime Cuts Nursery, paste the server URL, choose No authentication.
3. Click Scan tools, then Create.
4. Turn the app on in a conversation from the + menu.

Gemini Spark — Google AI Pro or Ultra, personal Google account, US, English

1. Go to gemini.google.com → Settings & help → Connected Apps.
2. Under Custom apps for Spark, click Add a custom app and paste the server URL.
3. Click Next, accept the connection notice, then Connect. Gemini lists the tools it synced.
4. Start a prompt with @Prime Cuts Nursery Public Data to use it.

Verified live 2026-09-08: Gemini synced all tools and quoted current stock and the seasonal per-plant price. The regular Gemini chat and Google AI Overviews do not call third-party tools; only Spark does.

Pro / Max, any model

1. Settings → Connectors → + Custom connector → Remote.
2. Name: Prime Cuts Nursery. URL: the server URL. Transport: Streamable HTTP. Auth: none.
3. Acknowledge the third-party notice and Add. One connector serves every model Perplexity runs.

any plan

1. Click the install button, or add the JSON below to ~/.cursor/mcp.json (or your project's .cursor/mcp.json).
2. Cursor lists the server under Settings → MCP; tools are available to the agent immediately.

{
  "mcpServers": {
    "prime-cuts-nursery": { "url": "https://www.primecutsnursery.com/api/mcp" }
  }
}

Copilot Chat agent mode

1. Click the install button, or add the JSON below to .vscode/mcp.json (root key is servers).
2. Open Copilot Chat in agent mode; the tools appear in the tool picker.

{
  "servers": {
    "prime-cuts-nursery": { "type": "http", "url": "https://www.primecutsnursery.com/api/mcp" }
  }
}

terminal

1. Run the command below once; it is saved for future sessions.

claude mcp add --transport http prime-cuts-nursery https://www.primecutsnursery.com/api/mcp

Block's open-source agent

1. Click the install button, or Extensions → Add custom extension → Streamable HTTP → paste the server URL.

local and self-hosted assistants

1. Add a remote MCP server of type Streamable HTTP with the server URL and no credentials.
2. Any model running inside an MCP-capable app can then read live inventory — including fully local models.

OpenAI, Anthropic, Gemini or Vertex AI SDKs; LangChain, Agents SDK, ADK

1. Point your framework's remote-MCP tool at the server URL, or load the OpenAPI 3.1 document for function calling over the REST API.
2. Both return identical JSON with an html_url to cite. No key, open CORS, 60 requests/minute per IP.

# OpenAI Responses API
tools=[{"type": "mcp", "server_label": "prime_cuts_nursery",
        "server_url": "https://www.primecutsnursery.com/api/mcp", "require_approval": "never"}]

# OpenAPI for function calling / custom GPT actions
https://www.primecutsnursery.com/openapi.json