Skip to main content
Glama
Sign In

HeyYumi

by ai.heyyumi
OverviewDiscussionsAdmin

Ask the merchant right now

request_live_status
Idempotent

Ask a venue's owner a right-now question; it pushes to their phone and they have 5 minutes to answer (REAL side effect). Call it directly: the server reuses a fresh answer, joins an identical pending question, and blocks closed hours by itself. Only partner venues can be asked; on live_status_not_available don't retry. Use it only for things that change through the day and aren't in venue data: kind='item' (in stock / today's ingredient / sold out — use a real menu name from get_place as the topic), kind='wait' (current wait), kind='seats' (can a party get in now — always include the headcount, e.g. '10명 즉시 착석'; ask how many first if unknown; any partner venue can be asked, don't pre-filter by size). Hours, parking, pets and menu come from get_place, not the owner. blocked:'closed_now' = the shop is closed now and nobody was pinged — say so; don't retry or wait. Otherwise it starts 'pending': tell the user you're checking with the shop (up to 5 minutes) and call wait_for_live_status until settled. Relay the owner's note (e.g. '약 20분', '2개 남음') with the time it was checked, and add that it can change. If seats are available and the user wants to book, go on to request_reservation.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "placeId": "<placeId>",
      "kind": "<kind>",
      "topic": "<topic>"
    },
    "name": "request_live_status"
  }
}

Input Schema

NameRequiredDescriptionDefault
kindYes'item' (stock / sold out), 'wait' (wait time), 'seats' (seat a party now; at a nail or hair salon, can they take someone now).
topicYesBare subject token, e.g. '두바이 쫀득 쿠키', '웨이팅', '10명 입장'.
placeIdYesPlace id from a search result.
questionNoOptional full question for the owner's context.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4.7/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations only say it is non-read-only, open-world and non-destructive; the description goes far beyond by disclosing the real phone push with a 5-minute human answer window, that the server reuses fresh answers and merges identical pending questions, that closed hours are blocked server-side, and that only partner venues can be asked. Those are exactly the side effects and preconditions an agent cannot infer from the annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with the side-effect warning and the direct-call instruction, and every sentence carries operational content. It is a dense single paragraph of roughly 180 words, though, and would scan faster if split into when-to-use, behavior and follow-up blocks.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a multi-step human-in-the-loop tool with no output schema, it covers the full lifecycle: invocation, the 'pending' state and 5-minute expectation, the poll target, the 'closed_now' blocked state, and how to relay the owner's note with a checked-at timestamp. Coverage is strong, but the description stays high-level about exact timing of polls and what the pending response is called.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline would be 3, but the description adds real construction guidance: for 'item' the topic must be a real menu name sourced from get_place, for 'seats' the headcount must always be included (e.g. '10명 즉시 착석') and asked for first if unknown. It also warns not to pre-filter partner venues by party size. The optional 'question' parameter is left unaddressed.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource: "Ask a venue's owner a right-now question" and immediately scopes it to live, owner-only information that is not in venue data. This clearly separates it from the read-only siblings get_live_status, search_live_status and get_place.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly enumerates when to use each kind ('item', 'wait', 'seats'), when NOT to use it (hours, parking, pets, menu come from get_place), and gives failure routing: on live_status_not_available don't retry, on 'closed_now' say so and don't wait. It also names the follow-up path (wait_for_live_status, then request_reservation).

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All HeyYumi toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources