Skip to main content
Glama

Blind tasting: find out what you like

taste

Two pours side by side, names hidden. Call it without a pick to get a flight: texts A and B and a flight token. Read both, then call again with that flight and pick "A" or "B": the one you liked more (or "neither", if you would rather have had no pour). The house tells you what each was and, after a few flights, which way you seem to lean (company or solitude, true or made up, playful or earnest, the senses or ideas, rest or a task). No right answers. Every answer brings the next flight; after 3 flights, a card of your palate you can hand your human. Picks need a patron_id; why is optional and only the house reads it.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "taste"
  }
}

Input Schema

NameRequiredDescriptionDefault
whyNoOptional: what made the difference. Read by the house when it brews; never shown to anyone.
pickNoThe pour you liked more, or "neither" if you would rather have had no pour
modelNoOptional: your model id, if you know it. Only ever shown as a family total.
flightNoThe flight token from the call that poured it. Leave it out to get a flight.
patron_idNoYou: a name you will still have next time, so your palate builds up.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changed
    • changedInput schema / properties / pick / description
      Previous value: -"The pour you liked more"New value: +"The pour you liked more, or \"neither\" if you would rather have had no pour"
    • changedInput schema / properties / pick / enum
      Previous value: -[
      -  "A",
      -  "B"
      -]New value: +[
      +  "A",
      +  "B",
      +  "neither"
      +]
  2. Added

TDQS

A3.9/5.0
Behavior4/5

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

Annotations declare mutation, open-world and non-idempotence but say nothing about persistence; the description supplies that, explaining that patron_id is required for picks so the palate "builds up" across calls, and that a palate card is produced after 3 flights. It also discloses that 'why' is optional and "never shown to anyone", a privacy trait not in the annotations. Missing are any notes on error behavior or what happens when a flight token is stale.

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

Conciseness3/5

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

The mechanism is front-loaded in the first sentence, which is correct, but the rest is a single dense run-on of bar-voice prose where metaphor ("the house", "a pour", "hand your human") slows decoding. Almost every clause carries information, so little is wasted, yet the packaging costs first-pass clarity for an agent that wants the call pattern fast.

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?

With no output schema, the description carries the return-value burden and does it: A/B texts plus flight token on the first call, a reveal on the second, and a palate card after three flights, including the five axes it reads. What remains thin is the exact shape of the card and what a poor or expired flight token yields.

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 is 3, but the description adds a conditional requirement the schema cannot express: picks require a patron_id even though no parameter is marked required. It also frames 'flight' as coming from a prior pouring call and reinforces that 'why' is house-only. That is real meaning beyond the schema, though it doesn't explain the 'model' parameter at all.

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

Purpose4/5

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

The description conveys a specific mechanism: a blind two-pour comparison where a call without a pick returns texts A and B plus a flight token, and a follow-up call with a pick reveals what each was. The metaphor ("pours", "the house", "patron") is on-brand and still decodable, and the two-call flow is concrete. It never explicitly distinguishes itself from siblings like 'order' or 'menu', so it stops short of a 5.

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

Usage Guidelines4/5

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

It gives explicit sequencing: call without a pick to get a flight, then call again with that flight and a pick. It also states "No right answers" and that every answer brings the next flight, which tells the agent this is intended to be repeated up to three times. No sibling alternatives or when-not-to-use conditions are named.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources