Sign inTry it

Find Italian recipes on GialloZafferano.

Searches GialloZafferano, Italy's largest recipe site: recipe title, course, rating out of 5, comments, difficulty, total time (as shown and in minutes), calories per portion and link, 15 per page with the total. Optional difficulty filter ('facile' includes 'molto facile'). A query with no recipe at all is redirected to the site-wide search, so the run fails instead of returning an empty list.

No reports yetRead-only, no login
Site
GialloZafferano, Italy
Version
v1
Runs, last 7 days
No reports yet
sha256
d751276816a5…

Run it with your agent

after installing Wield

Ask in plain words. Your agent finds this recipe and calls it:

“Find Italian recipes on GialloZafferano.”

Read-only: it changes nothing on the site. No login needed.

{
  "tool": "run",
  "arguments": {
    "recipe_id": "giallozafferano.it/search-recipes",
    "params": {
      "query": "lasagne"
    }
  }
}

Requests

sent from your browser, in order
  1. 1GET/ricerca-ricette/{{f.prefix}}{{params.query | trim | lower | urlencode}}/?difficolta[difficile]&difficolta[facile]&difficolta[media]GialloZafferano's search list (/ricerca-ricette/[pageN/]<query>/ with difficulty filters). Without a filter, page 1 is a set of themed carousels; with the difficulty filter (all three levels for 'any', which keeps every result) it is the plain list of article.gz-card-search cards.repeats, max 1

Parameters

from your agent
difficultyenum "any" | "facile" | "media" | "difficile" = "any"
pageinteger = 1
querystringrequiredPiatto o ingredienti, e.g. 'lasagne', 'tiramisù', 'pollo al forno'

Returns

to your agent
pageinteger
querystring
recipesobject[]comments, course, difficulty, kcal, rating, title, total_minutes, total_time, url
totalinteger

Where it can send

  • www.giallozafferano.it

Anything else is blocked by the extension.

Reads from your session

Nothing. Your browser attaches the site's cookies by itself.

Used only in requests to the addresses on the left. Never sent to wield.run or your agent.

Checks

a run passes when
  • length(output.recipes) == `15`
  • output.total > `50`
  • length(output.recipes[?starts_with(url, 'https://ricette.giallozafferano.it/')]) == length(output.recipes)
  • length(output.recipes[?rating != `null` && rating <= `5`]) > `10`
  • length(output.recipes[?difficulty != `null` && total_minutes != `null`]) > `10`

Raw recipe

Served as canonical JSON from /api/v1/recipes/giallozafferano.it/search-recipes, so its sha256 matches the ETag and the index.

Check it yourself
curl -s https://wield.run/api/v1/recipes/giallozafferano.it/search-recipes | shasum -a 256
# d751276816a5d24f72964a015d1c8ce5bd589f85223f94fae35642a9b20870f5
giallozafferano.it/search-recipes.json, v1
{
  "description": "Searches GialloZafferano, Italy's largest recipe site: recipe title, course, rating out of 5, comments, difficulty, total time (as shown and in minutes), calories per portion and link, 15 per page with the total. Optional difficulty filter ('facile' includes 'molto facile'). A query with no recipe at all is redirected to the site-wide search, so the run fails instead of returning an empty list.",
  "effect": "read",
  "id": "giallozafferano.it/search-recipes",
  "keywords": [
    "giallozafferano",
    "ricette",
    "ricetta",
    "cucina italiana",
    "primi piatti",
    "dolci",
    "lasagne",
    "pasta",
    "facile",
    "veloce",
    "recipes",
    "italian cooking",
    "recipe"
  ],
  "output": {
    "schema": {
      "properties": {
        "page": {
          "type": "integer"
        },
        "query": {
          "type": "string"
        },
        "recipes": {
          "items": {
            "properties": {
              "comments": {
                "type": "integer"
              },
              "course": {
                "description": "e.g. Primi piatti, Dolci",
                "type": [
                  "string",
                  "null"
                ]
              },
              "difficulty": {
                "description": "Molto facile, Facile, Media, Difficile",
                "type": [
                  "string",
                  "null"
                ]
              },
              "kcal": {
                "description": "Per portion",
                "type": [
                  "integer",
                  "null"
                ]
              },
              "rating": {
                "description": "0–5",
                "type": [
                  "number",
                  "null"
                ]
              },
              "title": {
                "type": "string"
              },
              "total_minutes": {
                "type": [
                  "integer",
                  "null"
                ]
              },
              "total_time": {
                "description": "As shown, e.g. '1 h 10 min'",
                "type": [
                  "string",
                  "null"
                ]
              },
              "url": {
                "type": "string"
              }
            },
            "type": "object"
          },
          "type": "array"
        },
        "total": {
          "type": [
            "integer",
            "null"
          ]
        }
      },
      "required": [
        "query",
        "recipes"
      ],
      "type": "object"
    },
    "value": "let $s = steps.search[0] in {query: params.query, total: ($s.count && to_number(replace(split($s.count, ' ')[0], '.', ''))) || `0`, page: params.page || `1`, recipes: map(&(let $time = (time && !starts_with(time, 'Kcal') && time) || '' in {title: title, course: course || `null`, rating: (rating && to_number(replace(rating, ',', '.'))) || `null`, comments: (comments && to_number(replace(comments, '.', ''))) || `0`, difficulty: difficulty || `null`, total_time: ($time != '' && $time) || `null`, total_minutes: let $w = split($time, ' ') in (length($w) == `2` && $w[1] == 'min' && to_number($w[0])) || (length($w) == `2` && $w[1] == 'h' && to_number($w[0]) * `60`) || (length($w) == `4` && $w[1] == 'h' && $w[3] == 'min' && to_number($w[0]) * `60` + to_number($w[2])) || `null`, kcal: (kcal && starts_with(kcal, 'Kcal') && to_number(split(kcal, ' ')[1])) || (time && starts_with(time, 'Kcal') && to_number(split(time, ' ')[1])) || `null`, url: href}), $s.recipes[?href != `null`] || `[]`)}"
  },
  "params": {
    "additionalProperties": false,
    "properties": {
      "difficulty": {
        "default": "any",
        "enum": [
          "any",
          "facile",
          "media",
          "difficile"
        ],
        "type": "string"
      },
      "page": {
        "default": 1,
        "maximum": 20,
        "minimum": 1,
        "type": "integer"
      },
      "query": {
        "description": "Piatto o ingredienti, e.g. 'lasagne', 'tiramisù', 'pollo al forno'",
        "maxLength": 100,
        "minLength": 2,
        "type": "string"
      }
    },
    "required": [
      "query"
    ],
    "type": "object"
  },
  "provenance": {
    "method": "browser-inspect",
    "notes": "GialloZafferano's search /ricerca-ricette/<query>/ shows themed carousels on page 1 (best results, most commented, light, plus blog and Cookaround picks); /ricerca-ricette/page1/… redirects there. Any filter turns page 1 into the plain list (article.gz-card-search, data-index 1–15; page N is /ricerca-ricette/pageN/<query>/). Selecting all three difficulty filters (difficolta[facile|media|difficile]=1, the page's own filter links) keeps every result (105 for lasagne, as unfiltered) and gives the list. Each card: h2.gz-title a (link to ricette.giallozafferano.it, never called), .gz-category, ul.gz-card-data.top (comments, rating '4,3'), ul.gz-card-data.bottom (difficulty, time '1 h 10 min', 'Kcal 719'). The total is .gz-custom-title > span.orange ('105 ricette'). Special editorial cards are not gz-card-search and are skipped. A query without recipes answers 301 to /ricerca/<query>/ (site-wide search), which the executor does not follow.",
    "recorded_at": "2026-11-07T22:49:00+01:00",
    "recorded_by": "C4"
  },
  "requires_login": false,
  "schema": "https://wield.run/schema/recipe/v1.json",
  "site": {
    "country": "IT",
    "id": "giallozafferano.it",
    "language": "it",
    "name": "GialloZafferano",
    "origin": "https://www.giallozafferano.it",
    "origins": [
      "https://www.giallozafferano.it"
    ]
  },
  "steps": [
    {
      "description": "GialloZafferano's search list (/ricerca-ricette/[pageN/]<query>/ with difficulty filters). Without a filter, page 1 is a set of themed carousels; with the difficulty filter (all three levels for 'any', which keeps every result) it is the plain list of article.gz-card-search cards.",
      "expect": {
        "content_type": "html",
        "status": [
          200
        ]
      },
      "extract": {
        "html": {
          "count": {
            "selector": ".gz-custom-title > span.orange"
          }
        },
        "html_items": {
          "recipes": {
            "fields": {
              "comments": {
                "selector": "ul.gz-card-data.top li:nth-child(1)"
              },
              "course": {
                "selector": ".gz-category"
              },
              "difficulty": {
                "selector": "ul.gz-card-data.bottom li:nth-child(1)"
              },
              "href": {
                "attr": "href",
                "selector": "h2.gz-title a"
              },
              "kcal": {
                "selector": "ul.gz-card-data.bottom li:nth-child(3)"
              },
              "rating": {
                "selector": "ul.gz-card-data.top li:nth-child(2)"
              },
              "time": {
                "selector": "ul.gz-card-data.bottom li:nth-child(2)"
              },
              "title": {
                "selector": "h2.gz-title a"
              }
            },
            "max": 15,
            "selector": "article.gz-card-search"
          }
        }
      },
      "foreach": {
        "as": "f",
        "in": "let $d = params.difficulty || 'any', $p = params.page || `1` in [{prefix: ($p > `1` && join('', ['page', to_string($p), '/'])) || '', facile: ($d == 'any' || $d == 'facile') && '1' || `null`, media: ($d == 'any' || $d == 'media') && '1' || `null`, difficile: ($d == 'any' || $d == 'difficile') && '1' || `null`}]",
        "max": 1
      },
      "id": "search",
      "request": {
        "headers": {
          "Accept": "text/html"
        },
        "method": "GET",
        "query": {
          "difficolta[difficile]": "{{f.difficile}}",
          "difficolta[facile]": "{{f.facile}}",
          "difficolta[media]": "{{f.media}}"
        },
        "url": "/ricerca-ricette/{{f.prefix}}{{params.query | trim | lower | urlencode}}/"
      }
    }
  ],
  "task": "Find Italian recipes on GialloZafferano",
  "tos": {
    "notes": "GialloZafferano's robots.txt disallows the whole site for Claude-Web and anthropic-ai (Disallow: /); its rules for other agents allow the search. Wield keeps such sites by the maintainers' decision, with this note. Scope: the recipe runs only when a user asks, makes one search request per call from the user's own browser, reads one result page, and never crawls recipe pages or collects in bulk.",
    "reviewed": false
  },
  "verify": {
    "checks": [
      "length(output.recipes) == `15`",
      "output.total > `50`",
      "length(output.recipes[?starts_with(url, 'https://ricette.giallozafferano.it/')]) == length(output.recipes)",
      "length(output.recipes[?rating != `null` && rating <= `5`]) > `10`",
      "length(output.recipes[?difficulty != `null` && total_minutes != `null`]) > `10`"
    ],
    "example_params": {
      "query": "lasagne"
    },
    "max_ms": 4000
  },
  "version": 1
}