Skip to main content
Glama
Kitchen Pass MCP server

Kitchen Pass MCP server

Official

Whole-menu review: what to fix first (paid, $1)

review_menu
Destructive

Review an entire menu at once to calculate plate costs, food cost percentages, menu-engineering classes, and rank dishes by weekly margin shortfall so you know what to fix first.

Instructions

Whole-menu review from head chef Cristopher's verified costing rules. Send every dish: menu price and ingredient lines or plate cost, optional section and a week of units sold. Get plate cost with his trim and cook loss on brisket, lamb leg, salmon and 30+ more, food cost %, on target or over, price at target, Star, Plowhorse, Puzzle or Dog, and fix these first ranked by weekly margin shortfall, with re-cost advice and workings. Only a 200 answer is charged; refusals and 404s are free.

Use when the user wants their whole menu reviewed at once. Send 2-200 dishes, each with name, menuPrice (as shown on the menu) and either lines or plateCost; add unitsSold (one week) for every dish or none, for classes and the fix-first ranking. For one dish only, use calculate_menu_price; for classes from plate costs alone, calculate_menu_engineering costs less. Pass on each dish's action and price advice as given. Costs $1. Paid automatically in USDC with x402 (test USDC when this server is set to base-sepolia), within this server's per-session and per-day spend caps. Only a 200 answer is charged; 400, 404 and 422 are free. Set dryRun to true to check for free whether this would be answered and what it costs; nothing is signed. If unsure what is covered, call check_coverage first (free). Not every answer is chef-verified: say "chef-verified by Cristopher" only when the result's label says exactly that; otherwise say it is not chef-verified. A refusal is final: don't rephrase the question to get around it.

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": {
      "dishes": "<dishes>"
    },
    "name": "review_menu"
  }
}

Input Schema

NameRequiredDescriptionDefault
dishesYes
dryRunNoFree quote only: check whether this would be answered and what it costs. Nothing is signed or charged.
taxPctNoOnly with pricesIncludeTax true.
roundToNo
targetPctNo
allowancePctNo
pricesIncludeTaxNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A4.6/5.0
Behavior5/5

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

Beyond the annotations (destructiveHint/openWorldHint), the description discloses the commercial terms an agent most needs: $1 cost, charging only on a 200 response, free 400/404/422, automatic USDC payment via x402 including the base-sepolia test-USDC caveat, and per-session/per-day spend caps. It also explains that dryRun signs and charges nothing, and adds a critical output caveat that 'chef-verified by Cristopher' may only be claimed when the result label says so.

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

Conciseness5/5

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

It is dense but front-loaded: outputs first, then usage and input shape, then pricing/payment mechanics, then output caveats. The only real redundancy is the charging rule appearing twice ('Only a 200 answer is charged' near the top and again in the payment paragraph), which is minor against otherwise high-information sentences.

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 complex, paid, 7-parameter tool with no output schema, the definition covers purpose, input shape, output fields, charging model, and safe-preview path thoroughly, so an agent can call it correctly. The one gap is the pricing-adjustment knobs (targetPct, allowancePct, taxPct/pricesIncludeTax, roundTo), which are left entirely to a partially documented schema.

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

Parameters3/5

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

Schema description coverage is low (29%), and the description compensates for the highest-value part: the dish payload (name, menuPrice, lines vs plateCost, optional section, one week of unitsSold) and dryRun. However, it says nothing about taxPct, pricesIncludeTax, allowancePct, targetPct, or roundTo, which materially change the computed food cost and target price and are the parameters most likely to be mis-set.

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?

The title and opening sentence state a specific verb and resource (a whole-menu review) and enumerate the concrete outputs: plate cost with trim/cook loss, food cost %, on-target status, target price, Star/Plowhorse/Puzzle/Dog class, and a fix-first ranking by weekly margin shortfall. It also names the sibling it is not ('For one dish only, use calculate_menu_price'), so an agent can distinguish it without opening another schema.

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?

Guidance is explicit on when to use it ('when the user wants their whole menu reviewed at once'), the input envelope (2-200 dishes, name + menuPrice + either lines or plateCost), the all-or-none rule for unitsSold, and the alternative paths (calculate_menu_price for a single dish, calculate_menu_engineering for classes from plate costs alone). It further routes the agent to check_coverage when scope is unclear and to dryRun for a free quote, and it states that a refusal is final and must not be rephrased around.

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