Add real-time local restaurant data to any Flowise chatflow or agent flow. Drop in a Custom MCP node, point it at https://stayinloop.dev/mcp, and all four Loop tools appear automatically in your Agent node. No API key required on the free tier.
In Flowise, search the node panel for Custom MCPand drag it onto your canvas. Set the configuration to the JSON below. Flowise connects over HTTP and discovers Loop's four tools automatically.
{ "url": "https://stayinloop.dev/mcp" } // With an API key (higher-volume projects): { "url": "https://stayinloop.dev/mcp", "headers": { "Authorization": "Bearer <your-api-key>" } }
Connect the Custom MCP node's toolsoutput to your Agent node's Tools input. The agent calls search(), get_details(), verify(), and report() based on conversation context — no manual routing needed.
# Flowise agent flow — restaurant search # # Nodes: # 1. Chat Input # ↓ # 2. Agent (LLM: GPT-4o or Claude) # Tools: [Custom MCP — Loop] # System prompt: "You help users find restaurants in Kreuzberg. # Use search() first, then get_details() when # the user picks one. Call verify() if they ask # about current hours. Call report() after they visit." # ↓ # 3. Chat Output # # The agent decides which Loop tool to call based on conversation context. # No manual routing logic needed.
After the user acts on a result, the agent calls report() via the MCP node. This mutates the record's confidence score in Loop's database — future agents searching the same area get better data.
# After user visits the restaurant: # The agent calls report() via the MCP node with: { "result_token": "<token from get_details>", "outcome": "correct" // correct · wrong · booked · closed · other } # result_token expires after 30 minutes. # Each report() call updates the record's confidence score in Loop's DB.
| Tool | When to call | Key output |
|---|---|---|
search() | User asks for a restaurant | Up to 8 ranked records |
get_details() | User selects a result | Full record + result_token |
verify() | User asks about current status | Live observation + confidence |
report() | After user acts on result | Mutates record confidence |
Full schema: api.stayinloop.dev/v1/openapi.json
Search the node panel for Custom MCP, drag it onto your canvas, and set the config to {"url": "https://stayinloop.dev/mcp"}. Flowise discovers Loop's four tools automatically. Wire the tools output into an Agent node.
Yes. The Custom MCP node works in both modes. Chatflow is best for conversational restaurant search; agent flow is best for multi-step workflows that search, verify, and report outcomes in sequence.
The result_token is returned by get_details() and is required to call report(). It is a signed HMAC token that expires after 30 minutes. In Flowise, the agent holds it in its context window between the get_details call and the subsequent report call.
No. The free tier accepts unauthenticated connections to https://stayinloop.dev/mcp. For higher-volume projects, apply for a key at stayinloop.dev/#pricing and add it as an Authorization: Bearer <key> header in the Custom MCP node configuration.