mcp-recipes

One dish, as each source writes it

compare_recipes

Read-onlyIdempotent

Shows how each recipe source writes the same dish side by side, rescaling every version to the same servings so ingredients, quantities, units, and yields can be compared directly.

Instructions

Take a dish and show how each source writes it, side by side. Each source's closest match is read in full, and all of them can be rescaled to the same number of servings so the ingredient lists stand comparison. The answer states what differs and leaves it there: the quantities each asks for, what each measures in, what each yields, and which fields each source publishes at all. No version is ranked above another. Some sources carry reader ratings and some have no author and no rating by nature, so there is no score they share. When only one source answers, the answer names that source and shows its version, so a single version is read as the one it is.

Input Schema

Name	Required	Description
`dish`	Yes	The dish to compare, in any of the languages the sources publish in.
`sources`	No	Sources to compare, by id. Left out, every source is read, which is the point of this tool. Name two or three to put particular traditions side by side. The ids are the ones 'per_source' reports.
`sections`	No	Which parts of each version to return. Several full recipes is a lot of text.
`servings`	No	Rescale every version to this many, which is what makes the lists comparable.
`max_steps`	No	Steps to return per version.
`max_step_chars`	No	Characters kept per step. Raise it only if a step was cut mid-sentence.

Output Schema

Name	Required	Description
`dish`	Yes
`notes`	Yes
`versions`	Yes
`per_source`	Yes
`differences`	Yes	What differs between the versions, stated as fact. No version is ranked.

Schema Changelog

Changes observed during successful MCP inspections.

Changed39 schema fields changedv4.1.0Oct 30, 2026

addedOutput schema / properties / per_source / items / properties / names_the_dish

Added value: +{
+  "description": "How many of this source's rows in this answer carry the dish in their title, out of 'count'. A row is what the source returned for the words it was handed, and one of these indexes answers any wording with something, so zero means the source offered rows and none of them is the dish.",
+  "maximum": 9007199254740991,
+  "minimum": -9007199254740991,
+  "type": "integer"
+}

removedOutput schema / properties / per_source / items / properties / reported_total_means / anyOf
```
Removed value: -[
-  {
-    "type": "string"
-  },
-  {
-    "type": "null"
-  }
-]
```
addedOutput schema / properties / per_source / items / properties / reported_total_means / type
```
Added value: +[
+  "string",
+  "null"
+]
```
removedOutput schema / properties / per_source / items / properties / wordings / items / properties / not_run_because / anyOf
```
Removed value: -[
-  {
-    "type": "string"
-  },
-  {
-    "type": "null"
-  }
-]
```
addedOutput schema / properties / per_source / items / properties / wordings / items / properties / not_run_because / type
```
Added value: +[
+  "string",
+  "null"
+]
```

changedOutput schema / properties / per_source / items / required

Previous value: -[
-  "source",
-  "name",
-  "status",
-  "count",
-  "reported_total",
-  "reported_total_means",
-  "skipped",
-  "cached",
-  "preferred_by_name",
-  "wordings",
-  "error",
-  "read"
-]New value: +[
+  "source",
+  "name",
+  "status",
+  "count",
+  "reported_total",
+  "reported_total_means",
+  "skipped",
+  "cached",
+  "preferred_by_name",
+  "names_the_dish",
+  "wordings",
+  "error",
+  "read"
+]

removedOutput schema / properties / versions / items / properties / author / anyOf

Removed value: -[
-  {
-    "type": "string"
-  },
-  {
-    "type": "null"
-  }
-]

addedOutput schema / properties / versions / items / properties / author / type
```
Added value: +[
+  "string",
+  "null"
+]
```

removedOutput schema / properties / versions / items / properties / category / anyOf

Removed value: -[
-  {
-    "type": "string"
-  },
-  {
-    "type": "null"
-  }
-]

addedOutput schema / properties / versions / items / properties / category / type
```
Added value: +[
+  "string",
+  "null"
+]
```

removedOutput schema / properties / versions / items / properties / image_url / anyOf

Removed value: -[
-  {
-    "type": "string"
-  },
-  {
-    "type": "null"
-  }
-]

addedOutput schema / properties / versions / items / properties / image_url / type
```
Added value: +[
+  "string",
+  "null"
+]
```
removedOutput schema / properties / versions / items / properties / ingredients / items / properties / amount / anyOf
```
Removed value: -[
-  {
-    "type": "number"
-  },
-  {
-    "type": "null"
-  }
-]
```
addedOutput schema / properties / versions / items / properties / ingredients / items / properties / amount / type
```
Added value: +[
+  "number",
+  "null"
+]
```
removedOutput schema / properties / versions / items / properties / ingredients / items / properties / amount_max / anyOf
```
Removed value: -[
-  {
-    "type": "number"
-  },
-  {
-    "type": "null"
-  }
-]
```
addedOutput schema / properties / versions / items / properties / ingredients / items / properties / amount_max / type
```
Added value: +[
+  "number",
+  "null"
+]
```

addedOutput schema / properties / versions / items / properties / ingredients / items / properties / is_equipment

Added value: +{
+  "description": "True for a line naming a tool, which some sites write among the ingredients. Such a line is never multiplied: a recipe made for more people uses the same pan.",
+  "type": "boolean"
+}

changedOutput schema / properties / versions / items / properties / ingredients / items / properties / language / enum
```
Previous value: -[
-  "fr",
-  "en"
-]New value: +[
+  "fr",
+  "en",
+  "es"
+]
```

changedOutput schema / properties / versions / items / properties / ingredients / items / properties / scaling / description

Previous value: -"'scaled' means the arithmetic landed on the exact product, 'rounded' that something had to move for the line to stay usable or that the line carries a second quantity whose match is the page's claim, 'unscaled' that there is nothing to multiply."New value: +"'scaled' means the arithmetic landed on the exact product, 'rounded' that something had to move for the line to stay usable or that the line carries a second quantity whose match is the page's claim, 'unscaled' that the line was left as given. Read 'is_equipment' beside it: a tool line is 'unscaled' whether or not it carries a figure."

removedOutput schema / properties / versions / items / properties / ingredients / items / properties / unit / anyOf
```
Removed value: -[
-  {
-    "type": "string"
-  },
-  {
-    "type": "null"
-  }
-]
```
addedOutput schema / properties / versions / items / properties / ingredients / items / properties / unit / type
```
Added value: +[
+  "string",
+  "null"
+]
```

changedOutput schema / properties / versions / items / properties / ingredients / items / required

Previous value: -[
-  "text",
-  "original",
-  "scaling",
-  "amount",
-  "amount_max",
-  "unit",
-  "language"
-]New value: +[
+  "text",
+  "original",
+  "scaling",
+  "amount",
+  "amount_max",
+  "unit",
+  "language",
+  "is_equipment"
+]

changedOutput schema / properties / versions / items / properties / language / enum

Previous value: -[
-  "fr",
-  "en"
-]New value: +[
+  "fr",
+  "en",
+  "es"
+]

changedOutput schema / properties / versions / items / properties / rating / anyOf

Previous value: -[
-  {
-    "additionalProperties": false,
-    "properties": {
-      "count": {
-        "anyOf": [
-          {
-            "maximum": 9007199254740991,
-            "minimum": -9007199254740991,
-            "type": "integer"
-          },
-          {
-            "type": "null"
-          }
-        ],
-        "description": "How many readers rated it, when the source says."
-      },
-      "max": {
-        "anyOf": [
-          {
-            "type": "number"
-          },
-          {
-            "type": "null"
-          }
-        ],
-        "description": "The top of the scale. A rating read without its scale means nothing."
-      },
-      "value": {
-        "type": "number"
-      }
-    },
-    "required": [
-      "value",
-      "count",
-      "max"
-    ],
-    "type": "object"
-  },
-  {
-    "type": "null"
-  }
-]New value: +[
+  {
+    "additionalProperties": false,
+    "properties": {
+      "count": {
+        "anyOf": [
+          {
+            "maximum": 9007199254740991,
+            "minimum": -9007199254740991,
+            "type": "integer"
+          },
+          {
+            "type": "null"
+          }
+        ],
+        "description": "How many readers rated it, when the source says."
+      },
+      "max": {
+        "description": "The top of the scale. A rating read without its scale means nothing.",
+        "type": [
+          "number",
+          "null"
+        ]
+      },
+      "value": {
+        "type": "number"
+      }
+    },
+    "required": [
+      "value",
+      "count",
+      "max"
+    ],
+    "type": "object"
+  },
+  {
+    "type": "null"
+  }
+]

addedOutput schema / properties / versions / items / properties / scaling_summary / description
```
Added value: +"The four counts add up to the number of ingredient lines."
```

addedOutput schema / properties / versions / items / properties / scaling_summary / properties / equipment_count

Added value: +{
+  "description": "Lines naming a tool, which some sites write among the ingredients.",
+  "maximum": 9007199254740991,
+  "minimum": -9007199254740991,
+  "type": "integer"
+}

changedOutput schema / properties / versions / items / properties / scaling_summary / required

Previous value: -[
-  "scaled_count",
-  "rounded_count",
-  "unscaled_count"
-]New value: +[
+  "scaled_count",
+  "rounded_count",
+  "unscaled_count",
+  "equipment_count"
+]

removedOutput schema / properties / versions / items / properties / steps_as_one_block / anyOf
```
Removed value: -[
-  {
-    "type": "boolean"
-  },
-  {
-    "type": "null"
-  }
-]
```
addedOutput schema / properties / versions / items / properties / steps_as_one_block / type
```
Added value: +[
+  "boolean",
+  "null"
+]
```
removedOutput schema / properties / versions / items / properties / yield / properties / factor / anyOf
```
Removed value: -[
-  {
-    "type": "number"
-  },
-  {
-    "type": "null"
-  }
-]
```
addedOutput schema / properties / versions / items / properties / yield / properties / factor / type
```
Added value: +[
+  "number",
+  "null"
+]
```
removedOutput schema / properties / versions / items / properties / yield / properties / original_count / anyOf
```
Removed value: -[
-  {
-    "type": "number"
-  },
-  {
-    "type": "null"
-  }
-]
```
addedOutput schema / properties / versions / items / properties / yield / properties / original_count / type
```
Added value: +[
+  "number",
+  "null"
+]
```
removedOutput schema / properties / versions / items / properties / yield / properties / original_max / anyOf
```
Removed value: -[
-  {
-    "type": "number"
-  },
-  {
-    "type": "null"
-  }
-]
```
addedOutput schema / properties / versions / items / properties / yield / properties / original_max / type
```
Added value: +[
+  "number",
+  "null"
+]
```
removedOutput schema / properties / versions / items / properties / yield / properties / original_text / anyOf
```
Removed value: -[
-  {
-    "type": "string"
-  },
-  {
-    "type": "null"
-  }
-]
```
addedOutput schema / properties / versions / items / properties / yield / properties / original_text / type
```
Added value: +[
+  "string",
+  "null"
+]
```
removedOutput schema / properties / versions / items / properties / yield / properties / unit / anyOf
```
Removed value: -[
-  {
-    "type": "string"
-  },
-  {
-    "type": "null"
-  }
-]
```
addedOutput schema / properties / versions / items / properties / yield / properties / unit / type
```
Added value: +[
+  "string",
+  "null"
+]
```

Changed5 schema fields changedv3.0.1Aug 29, 2026

addedInput schema / properties / sources

Added value: +{
+  "description": "Sources to compare, by id. Left out, every source is read, which is the point of this tool. Name two or three to put particular traditions side by side. The ids are the ones 'per_source' reports.",
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}

addedOutput schema / properties / versions / items / properties / rest_minutes

Added value: +{
+  "anyOf": [
+    {
+      "maximum": 9007199254740991,
+      "minimum": -9007199254740991,
+      "type": "integer"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "description": "Time the recipe stands, from a source that publishes it apart. It is in no other time here, and belongs to no other source's cooking time. Null where the source publishes none, which is not a dish that needs no rest."
+}

addedOutput schema / properties / versions / items / properties / steps_as_one_block

Added value: +{
+  "anyOf": [
+    {
+      "type": "boolean"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "description": "Whether the source publishes the method as one block of prose rather than as steps of its own. Null where the source says neither, so a single entry is not read as step one of several."
+}

addedOutput schema / properties / versions / items / properties / withheld

Added value: +{
+  "anyOf": [
+    {
+      "additionalProperties": false,
+      "properties": {
+        "parts": {
+          "description": "Which parts, named as 'ingredients' or 'method'.",
+          "items": {
+            "type": "string"
+          },
+          "type": "array"
+        },
+        "why": {
+          "description": "What the source says about keeping them back.",
+          "type": "string"
+        }
+      },
+      "required": [
+        "parts",
+        "why"
+      ],
+      "type": "object"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "description": "A part the source published nothing of because it keeps it for its subscribers. Null from a source that withholds nothing, which is a different statement from an empty list."
+}

changedOutput schema / properties / versions / items / required

Previous value: -[
-  "id",
-  "source",
-  "source_name",
-  "language",
-  "title",
-  "url",
-  "image_url",
-  "yield",
-  "ingredients",
-  "steps",
-  "prep_minutes",
-  "cook_minutes",
-  "total_minutes",
-  "category",
-  "author",
-  "rating",
-  "nutrition",
-  "equipment",
-  "tips",
-  "license",
-  "attribution",
-  "sections_returned",
-  "sections_omitted",
-  "scaling_summary"
-]New value: +[
+  "id",
+  "source",
+  "source_name",
+  "language",
+  "title",
+  "url",
+  "image_url",
+  "yield",
+  "ingredients",
+  "steps",
+  "prep_minutes",
+  "cook_minutes",
+  "total_minutes",
+  "rest_minutes",
+  "steps_as_one_block",
+  "withheld",
+  "category",
+  "author",
+  "rating",
+  "nutrition",
+  "equipment",
+  "tips",
+  "license",
+  "attribution",
+  "sections_returned",
+  "sections_omitted",
+  "scaling_summary"
+]

Changed4 schema fields changedv1.3.0Aug 11, 2026

addedOutput schema / properties / per_source / items / properties / preferred_by_name

Added value: +{
+  "description": "Whether this source's rows were arranged so the ones whose title names the dish come first. An order over one source's own rows, never a score against another source.",
+  "type": "boolean"
+}

addedOutput schema / properties / per_source / items / properties / read

Added value: +{
+  "anyOf": [
+    {
+      "additionalProperties": false,
+      "properties": {
+        "error": {
+          "anyOf": [
+            {
+              "additionalProperties": false,
+              "properties": {
+                "code": {
+                  "type": "string"
+                },
+                "hint": {
+                  "type": "string"
+                },
+                "message": {
+                  "type": "string"
+                }
+              },
+              "required": [
+                "code",
+                "message"
+              ],
+              "type": "object"
+            },
+            {
+              "type": "null"
+            }
+          ]
+        },
+        "status": {
+          "enum": [
+            "read",
+            "failed"
+          ],
+          "type": "string"
+        }
+      },
+      "required": [
+        "status",
+        "error"
+      ],
+      "type": "object"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "description": "What became of the row this source offered, once it was opened. Null when no row was opened, which is what a search that did not answer and a search holding nothing both come to. A search and a read are two moments, and this is the second one."
+}

addedOutput schema / properties / per_source / items / properties / wordings

Added value: +{
+  "description": "Every wording this source was sent or held back from, in the order tried. A question written as a sentence is also sent in wordings derived from it, because these indexes answer the words they are handed. Counts here belong to one wording each and are never added.",
+  "items": {
+    "additionalProperties": false,
+    "properties": {
+      "added": {
+        "anyOf": [
+          {
+            "maximum": 9007199254740991,
+            "minimum": -9007199254740991,
+            "type": "integer"
+          },
+          {
+            "type": "null"
+          }
+        ],
+        "description": "Rows it returned that no earlier wording had."
+      },
+      "count": {
+        "anyOf": [
+          {
+            "maximum": 9007199254740991,
+            "minimum": -9007199254740991,
+            "type": "integer"
+          },
+          {
+            "type": "null"
+          }
+        ],
+        "description": "Rows this wording returned. Null when it did not run."
+      },
+      "derivation": {
+        "description": "How this wording was derived from the question.",
+        "type": "string"
+      },
+      "error": {
+        "anyOf": [
+          {
+            "additionalProperties": false,
+            "properties": {
+              "code": {
+                "type": "string"
+              },
+              "hint": {
+                "type": "string"
+              },
+              "message": {
+                "type": "string"
+              }
+            },
+            "required": [
+              "code",
+              "message"
+            ],
+            "type": "object"
+          },
+          {
+            "type": "null"
+          }
+        ]
+      },
+      "not_run_because": {
+        "anyOf": [
+          {
+            "type": "string"
+          },
+          {
+            "type": "null"
+          }
+        ],
+        "description": "Why it was held back. Null when it ran."
+      },
+      "query": {
+        "description": "What was sent to this source, or would have been.",
+        "type": "string"
+      },
+      "ran": {
+        "type": "boolean"
+      }
+    },
+    "required": [
+      "query",
+      "derivation",
+      "ran",
+      "count",
+      "added",
+      "not_run_because",
+      "error"
+    ],
+    "type": "object"
+  },
+  "type": "array"
+}

changedOutput schema / properties / per_source / items / required

Previous value: -[
-  "source",
-  "name",
-  "status",
-  "count",
-  "reported_total",
-  "reported_total_means",
-  "skipped",
-  "cached",
-  "error"
-]New value: +[
+  "source",
+  "name",
+  "status",
+  "count",
+  "reported_total",
+  "reported_total_means",
+  "skipped",
+  "cached",
+  "preferred_by_name",
+  "wordings",
+  "error",
+  "read"
+]

First observedv1.0.0Aug 7, 2026

TDQS

A4.1/5.0

Behavior4/5

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

Annotations already declare read-only, idempotent, open-world and non-destructive, so safety is covered. The description adds genuinely non-obvious behavior: no version is ranked, the sources share no comparable score, the result enumerates what differs and which fields each source publishes, and a lone source yields its version plainly. It doesn't touch cost/latency of reading several full recipes, though it does flag that sections defaults to ingredients because 'several full recipes is a lot of text.'

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?

Purpose is front-loaded and the sentences are ordered from what the tool does to how results behave. A few clauses are ornamental or circular, notably 'so a single version is read as the one it is,' which could be cut without loss.

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 an output schema present, the description needn't explain the return shape, and it still characterizes the response contents (quantities, units, yields, published fields) and the no-ranking contract. The single-source fallback is covered, so an agent has enough to call it confidently.

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 100%, so the schema already carries parameter meaning. The description restates the servings rescaling rationale, which the schema also states, and adds nothing about sources ids or section selection beyond what the properties document — a baseline 3.

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 opening sentence gives a concrete verb and resource — take a dish and show how each source writes it, side by side — which is clearly distinct from get_recipe (one version) and scale_ingredients (numeric scaling). An agent can route among the siblings on the strength of the first line alone.

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 explains the multi-source default ('Left out, every source is read, which is the point of this tool') and the narrowing case ('Name two or three to put particular traditions side by side'), plus the degenerate single-source outcome. It never explicitly names search_recipes or get_recipe as the alternative to pick instead, so the routing is implied rather than stated.

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

Deploy Server