Skip to main content
Glama
ashitamv

Recipe MCP

by ashitamv

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
PORTNoAccepted for managed hosting.
RECIPE_HOSTNoHTTP bind.127.0.0.1
RECIPE_MODENolive or demo.live
RECIPE_PORTNoHTTP bind port.8000
RECIPE_API_KEYNoTheMealDB development/educational key.1
RECIPE_RETRIESNoAdditional attempts.2
RECIPE_CACHE_TTLNoCache seconds.300
RECIPE_AUTH_TOKENNo32–256 visible ASCII characters; required outside loopback.
RECIPE_CACHE_SIZENoMaximum cache entries.128
RECIPE_ALLOWED_HOSTSNoJSON array of exact allowed hosts; :* permits ports.localhost and loopback
RECIPE_DATABASE_PATHNoDurable database path; use different files per mode.data/recipes.sqlite3
RECIPE_ALLOWED_ORIGINSNoJSON array of trusted origins; no global wildcard.localhost and loopback
RECIPE_REQUEST_TIMEOUTNoPer-attempt timeout.10

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Capabilities

Features and capabilities supported by this server

Protocol revision2025-11-25

CapabilityDetails
tools
{
  "listChanged": false
}
prompts
{
  "listChanged": false
}
resources
{
  "subscribe": false,
  "listChanged": false
}
experimental
{}

Tools

Functions exposed to the LLM to take actions

NameDescription
search_recipesB

Find recipes by dish name, save their details and return IDs and summaries.

search_by_first_letterB

Find and save recipes whose names begin with one ASCII letter.

get_recipe_detailsA

Read saved details or fetch and save a recipe by numeric ID.

get_random_recipeA

Fetch and save one random recipe. Each call can return a different result.

create_meal_planA

Atomically save a complete plan. Reuse the same key and payload when retrying.

Requires user confirmation. Unknown IDs fail the entire plan. A reused key with different content fails with idempotency_conflict.

get_meal_planA

Read a saved meal plan by its generated UUID.

generate_shopping_listA

Consolidate a saved plan's ingredients into a shopping list, without writes.

Uses current locally saved recipe details, once per unique recipe at its original batch size. Combines compatible quantities exactly. Show every unresolved measurement and warning to the user; never treat a partial known quantity as the complete amount. No serving-size scaling or regional cup/spoon conversions are inferred. No upstream API is called.

Prompts

Interactive templates invoked by user choice

NameDescription
generate_recipe_search_promptResearch dishes associated with a cuisine using grounded recipe data.
generate_meal_planning_promptDraft a plan, discuss constraints and save it after user confirmation.
generate_cooking_lesson_promptPrepare a cooking lesson with sourced examples and explicit learning goals.
generate_ingredient_exploration_promptCompare candidate recipes after checking their actual ingredient lists.
generate_cultural_cuisine_promptExplore cuisine with sourced recipes and qualified cultural context.

Resources

Contextual data attached and managed by the client

NameDescription
cuisinesList actual cuisines present in the local recipe collection.
meal_plansRead the latest 50 saved meal plans.
statisticsRead deduplicated recipe, cuisine and plan counts.

TDQS

A3.8/5.0

Scored across 7 tools

Disambiguation4/5

Most tools have clearly distinct purposes: recipe discovery (search, first-letter, random, by-ID), meal plan creation/reading, and shopping list generation. The only mild overlap is that multiple recipe tools all fetch and save recipe details, but their lookup modes are distinct enough to avoid serious misselection.

Naming Consistency5/5

All tool names follow a consistent snake_case verb-first pattern: create_, search_, get_, generate_. Even search_by_first_letter fits the convention with a descriptive prepositional modifier. There is no mixing of styles or vague verbs.

Tool Count5/5

Seven tools is a well-scoped size for a recipe/meal-planning server. Each tool covers a meaningful step in the workflow—recipe discovery, detail retrieval, plan creation/retrieval, and shopping list generation—without unnecessary redundancy.

Completeness4/5

The core workflow is covered end-to-end: find recipes, fetch details, create a meal plan, read it, and generate a shopping list. The main gaps are the lack of update/delete operations for saved recipes or meal plans and no way to list saved recipes, but these do not break the primary use case.

Maintenance

ActivityMaintained
ResponsivenessNo issues