{
  "openapi": "3.0.1",
  "info": {
    "title": "Sitemap URL Extractor: All URLs, Status Codes & New Pages",
    "description": "Get every URL of a website from its sitemaps: auto-discovery via robots.txt, nested indexes, .gz, RSS/Atom, lastmod filters, optional status-code checks and monitoring of new, changed and removed URLs. Failed sitemaps are free.",
    "version": "1.0",
    "x-build-id": "wiarrIJF8ggY2THMe"
  },
  "servers": [
    {
      "url": "https://api.apify.com/v2"
    }
  ],
  "paths": {
    "/acts/genesis.labs.co~sitemap-url-extractor-monitor/run-sync-get-dataset-items": {
      "post": {
        "operationId": "run-sync-get-dataset-items-genesis.labs.co-sitemap-url-extractor-monitor",
        "x-openai-isConsequential": false,
        "summary": "Executes an Actor, waits for its completion, and returns Actor's dataset items in response.",
        "tags": [
          "Run Actor"
        ],
        "requestBody": {
          "required": true,
          "content": {
            "application/json": {
              "schema": {
                "$ref": "#/components/schemas/inputSchema"
              }
            }
          }
        },
        "parameters": [
          {
            "name": "token",
            "in": "query",
            "required": true,
            "schema": {
              "type": "string"
            },
            "description": "Enter your Apify token here"
          }
        ],
        "responses": {
          "200": {
            "description": "OK"
          }
        }
      }
    },
    "/acts/genesis.labs.co~sitemap-url-extractor-monitor/runs": {
      "post": {
        "operationId": "runs-sync-genesis.labs.co-sitemap-url-extractor-monitor",
        "x-openai-isConsequential": false,
        "summary": "Executes an Actor and returns information about the initiated run in response.",
        "tags": [
          "Run Actor"
        ],
        "requestBody": {
          "required": true,
          "content": {
            "application/json": {
              "schema": {
                "$ref": "#/components/schemas/inputSchema"
              }
            }
          }
        },
        "parameters": [
          {
            "name": "token",
            "in": "query",
            "required": true,
            "schema": {
              "type": "string"
            },
            "description": "Enter your Apify token here"
          }
        ],
        "responses": {
          "200": {
            "description": "OK",
            "content": {
              "application/json": {
                "schema": {
                  "$ref": "#/components/schemas/runsResponseSchema"
                }
              }
            }
          }
        }
      }
    },
    "/acts/genesis.labs.co~sitemap-url-extractor-monitor/run-sync": {
      "post": {
        "operationId": "run-sync-genesis.labs.co-sitemap-url-extractor-monitor",
        "x-openai-isConsequential": false,
        "summary": "Executes an Actor, waits for completion, and returns the OUTPUT from Key-value store in response.",
        "tags": [
          "Run Actor"
        ],
        "requestBody": {
          "required": true,
          "content": {
            "application/json": {
              "schema": {
                "$ref": "#/components/schemas/inputSchema"
              }
            }
          }
        },
        "parameters": [
          {
            "name": "token",
            "in": "query",
            "required": true,
            "schema": {
              "type": "string"
            },
            "description": "Enter your Apify token here"
          }
        ],
        "responses": {
          "200": {
            "description": "OK"
          }
        }
      }
    }
  },
  "components": {
    "schemas": {
      "inputSchema": {
        "type": "object",
        "properties": {
          "startUrls": {
            "title": "Websites or sitemap URLs",
            "type": "array",
            "description": "Website addresses (example.com, https://example.com/blog/) or direct sitemap/feed URLs (https://example.com/sitemap_index.xml, .xml.gz, RSS/Atom, .txt). For a website, sitemaps are found via robots.txt and the usual locations. Leave empty to run a free demo.",
            "items": {
              "type": "object",
              "required": [
                "url"
              ],
              "properties": {
                "url": {
                  "type": "string",
                  "title": "URL of a web page",
                  "format": "uri"
                }
              }
            }
          },
          "urls": {
            "title": "Domains or URLs (plain text)",
            "type": "array",
            "description": "Same as above, as plain text: bare domains work here (example.com, shop.example.org). Handy for pasting a list or passing an array of strings from an integration.",
            "items": {
              "type": "string"
            }
          },
          "datasetId": {
            "title": "Input dataset (optional)",
            "type": "string",
            "description": "Read website/sitemap URLs from a dataset, e.g. the output of another Actor. In integrations pass {{resource.defaultDatasetId}}. Only this dataset is readable (limited permissions)."
          },
          "urlField": {
            "title": "URL field in the input dataset",
            "type": "string",
            "description": "Field holding the website or sitemap URL in each input row. Dot paths work, e.g. company.website.",
            "default": "url"
          },
          "maxInputItems": {
            "title": "Max input rows",
            "minimum": 0,
            "type": "integer",
            "description": "Read at most this many rows from the input dataset. 0 = all.",
            "default": 0
          },
          "discoverSitemaps": {
            "title": "Auto-discover sitemaps",
            "type": "boolean",
            "description": "For website URLs, look up Sitemap: lines in robots.txt, then /sitemap.xml, /sitemap_index.xml, /wp-sitemap.xml, /sitemap.xml.gz and /sitemap.txt. Turn off to treat every input URL as a sitemap file.",
            "default": true
          },
          "includeUrlPatterns": {
            "title": "Only URLs matching",
            "type": "array",
            "description": "Keep only URLs that match at least one pattern. Plain text matches anywhere (/blog/), * is a wildcard (https://example.com/products/*), or use /regex/.",
            "items": {
              "type": "string"
            }
          },
          "excludeUrlPatterns": {
            "title": "Skip URLs matching",
            "type": "array",
            "description": "Drop URLs that match any pattern (same syntax), e.g. /tag/ or *?replytocom=*.",
            "items": {
              "type": "string"
            }
          },
          "lastmodAfter": {
            "title": "Modified after",
            "type": "string",
            "description": "Keep URLs whose <lastmod> is on or after this date: 2026-01-31 or a relative time like \"7 days\" or \"24 hours\". Old child sitemaps are skipped entirely."
          },
          "lastmodBefore": {
            "title": "Modified before",
            "type": "string",
            "description": "Keep URLs whose <lastmod> is on or before this date (same formats). Useful to find stale pages."
          },
          "keepUrlsWithoutLastmod": {
            "title": "Keep URLs without lastmod",
            "type": "boolean",
            "description": "When a date filter is set, also keep URLs that have no <lastmod>.",
            "default": false
          },
          "sameSiteOnly": {
            "title": "Same site only",
            "type": "boolean",
            "description": "Drop URLs on other hostnames than the site (www. is ignored).",
            "default": false
          },
          "includeSitemapExtras": {
            "title": "Include image, video, news and hreflang data",
            "type": "boolean",
            "description": "Add image URLs, video info, Google News fields and hreflang alternates when the sitemap has them.",
            "default": true
          },
          "maxUrls": {
            "title": "Max URLs",
            "minimum": 0,
            "type": "integer",
            "description": "Stop after saving this many URLs in total. 0 = no limit (use Max cost per run to cap spend).",
            "default": 0
          },
          "maxUrlsPerSite": {
            "title": "Max URLs per site",
            "minimum": 0,
            "type": "integer",
            "description": "Per website or sitemap input. 0 = no limit.",
            "default": 0
          },
          "maxSitemapsPerSite": {
            "title": "Max sitemap files per site",
            "minimum": 1,
            "maximum": 100000,
            "type": "integer",
            "description": "Safety limit for huge sitemap indexes.",
            "default": 1000
          },
          "maxDepth": {
            "title": "Max index depth",
            "minimum": 0,
            "maximum": 20,
            "type": "integer",
            "description": "How many levels of nested sitemap indexes to follow.",
            "default": 5
          },
          "monitorMode": {
            "title": "Only new URLs (monitoring)",
            "enum": [
              "off",
              "new",
              "newAndChanged"
            ],
            "type": "string",
            "description": "Remembers each site's URLs between runs (in a key-value store named sitemap-monitor-<Monitor name> on your account) and outputs only what is new. Schedule it daily to get new pages, products or articles.",
            "default": "off"
          },
          "monitorName": {
            "title": "Monitor name",
            "type": "string",
            "description": "Separate memories for separate jobs, e.g. competitor-blogs. Letters, digits and dashes.",
            "default": "default"
          },
          "includeRemoved": {
            "title": "Also report removed URLs",
            "type": "boolean",
            "description": "Output URLs that were in the sitemap last time but are gone now (changeType \"removed\"). Only reported when the whole sitemap was read without errors or limits.",
            "default": false
          },
          "silentFirstRun": {
            "title": "Silent first run",
            "type": "boolean",
            "description": "On the first run of a monitor, just remember the URLs and output nothing (nothing charged per URL), so later runs return only real changes.",
            "default": false
          },
          "resetMonitor": {
            "title": "Reset memory",
            "type": "boolean",
            "description": "Forget the remembered URLs of these sites before the run (everything counts as new).",
            "default": false
          },
          "checkStatus": {
            "title": "Check status codes",
            "type": "boolean",
            "description": "Request every saved URL and add its HTTP status, final URL after redirects, redirect chain, content type, X-Robots-Tag and response time. Finds 404s and redirects listed in your sitemap.",
            "default": false
          },
          "checkMethod": {
            "title": "Status check method",
            "enum": [
              "HEAD",
              "GET"
            ],
            "type": "string",
            "description": "HEAD is fastest; servers that reject HEAD (405/501/403/404) are re-checked with GET automatically.",
            "default": "HEAD"
          },
          "onlyBroken": {
            "title": "Only output broken URLs",
            "type": "boolean",
            "description": "Save only URLs with status 4xx/5xx or a network error. All checks are still counted.",
            "default": false
          },
          "checkConcurrency": {
            "title": "Status check concurrency",
            "minimum": 1,
            "maximum": 50,
            "type": "integer",
            "description": "Parallel status checks.",
            "default": 10
          },
          "requestsPerSecond": {
            "title": "Requests per second",
            "minimum": 1,
            "maximum": 100,
            "type": "integer",
            "description": "Global politeness limit for all requests (sitemaps and status checks).",
            "default": 10
          },
          "timeoutSecs": {
            "title": "Request timeout (seconds)",
            "minimum": 5,
            "maximum": 180,
            "type": "integer",
            "description": "Per request. Timeouts and 5xx are retried twice.",
            "default": 30
          },
          "maxSitemapMB": {
            "title": "Max sitemap file size (MB)",
            "minimum": 1,
            "maximum": 500,
            "type": "integer",
            "description": "Larger files are skipped with TOO_LARGE (the sitemap protocol limit is 50 MB uncompressed).",
            "default": 60
          },
          "proxyConfiguration": {
            "title": "Proxy configuration",
            "type": "object",
            "description": "Optional. Use a proxy if a site blocks datacenter IPs. By default requests go out directly.",
            "default": {
              "useApifyProxy": false
            }
          },
          "failOnSiteError": {
            "title": "Fail the run on any error",
            "type": "boolean",
            "description": "By default the run succeeds when at least one sitemap was read, and problems are listed in the ERRORS record. Turn on to make integrations notice any failed sitemap.",
            "default": false
          }
        }
      },
      "runsResponseSchema": {
        "type": "object",
        "properties": {
          "data": {
            "type": "object",
            "properties": {
              "id": {
                "type": "string"
              },
              "actId": {
                "type": "string"
              },
              "userId": {
                "type": "string"
              },
              "startedAt": {
                "type": "string",
                "format": "date-time",
                "example": "2025-01-08T00:00:00.000Z"
              },
              "finishedAt": {
                "type": "string",
                "format": "date-time",
                "example": "2025-01-08T00:00:00.000Z"
              },
              "status": {
                "type": "string",
                "example": "READY"
              },
              "meta": {
                "type": "object",
                "properties": {
                  "origin": {
                    "type": "string",
                    "example": "API"
                  },
                  "userAgent": {
                    "type": "string"
                  }
                }
              },
              "stats": {
                "type": "object",
                "properties": {
                  "inputBodyLen": {
                    "type": "integer",
                    "example": 2000
                  },
                  "rebootCount": {
                    "type": "integer",
                    "example": 0
                  },
                  "restartCount": {
                    "type": "integer",
                    "example": 0
                  },
                  "resurrectCount": {
                    "type": "integer",
                    "example": 0
                  },
                  "computeUnits": {
                    "type": "integer",
                    "example": 0
                  }
                }
              },
              "options": {
                "type": "object",
                "properties": {
                  "build": {
                    "type": "string",
                    "example": "latest"
                  },
                  "timeoutSecs": {
                    "type": "integer",
                    "example": 300
                  },
                  "memoryMbytes": {
                    "type": "integer",
                    "example": 1024
                  },
                  "diskMbytes": {
                    "type": "integer",
                    "example": 2048
                  }
                }
              },
              "buildId": {
                "type": "string"
              },
              "defaultKeyValueStoreId": {
                "type": "string"
              },
              "defaultDatasetId": {
                "type": "string"
              },
              "defaultRequestQueueId": {
                "type": "string"
              },
              "buildNumber": {
                "type": "string",
                "example": "1.0.0"
              },
              "containerUrl": {
                "type": "string"
              },
              "usage": {
                "type": "object",
                "properties": {
                  "ACTOR_COMPUTE_UNITS": {
                    "type": "integer",
                    "example": 0
                  },
                  "DATASET_READS": {
                    "type": "integer",
                    "example": 0
                  },
                  "DATASET_WRITES": {
                    "type": "integer",
                    "example": 0
                  },
                  "KEY_VALUE_STORE_READS": {
                    "type": "integer",
                    "example": 0
                  },
                  "KEY_VALUE_STORE_WRITES": {
                    "type": "integer",
                    "example": 1
                  },
                  "KEY_VALUE_STORE_LISTS": {
                    "type": "integer",
                    "example": 0
                  },
                  "REQUEST_QUEUE_READS": {
                    "type": "integer",
                    "example": 0
                  },
                  "REQUEST_QUEUE_WRITES": {
                    "type": "integer",
                    "example": 0
                  },
                  "DATA_TRANSFER_INTERNAL_GBYTES": {
                    "type": "integer",
                    "example": 0
                  },
                  "DATA_TRANSFER_EXTERNAL_GBYTES": {
                    "type": "integer",
                    "example": 0
                  },
                  "PROXY_RESIDENTIAL_TRANSFER_GBYTES": {
                    "type": "integer",
                    "example": 0
                  },
                  "PROXY_SERPS": {
                    "type": "integer",
                    "example": 0
                  },
                  "PROXY_UNBLOCKER_UNITS": {
                    "type": "integer",
                    "example": 0
                  }
                }
              },
              "usageTotalUsd": {
                "type": "number",
                "example": 0.00005
              },
              "usageUsd": {
                "type": "object",
                "properties": {
                  "ACTOR_COMPUTE_UNITS": {
                    "type": "integer",
                    "example": 0
                  },
                  "DATASET_READS": {
                    "type": "integer",
                    "example": 0
                  },
                  "DATASET_WRITES": {
                    "type": "integer",
                    "example": 0
                  },
                  "KEY_VALUE_STORE_READS": {
                    "type": "integer",
                    "example": 0
                  },
                  "KEY_VALUE_STORE_WRITES": {
                    "type": "number",
                    "example": 0.00005
                  },
                  "KEY_VALUE_STORE_LISTS": {
                    "type": "integer",
                    "example": 0
                  },
                  "REQUEST_QUEUE_READS": {
                    "type": "integer",
                    "example": 0
                  },
                  "REQUEST_QUEUE_WRITES": {
                    "type": "integer",
                    "example": 0
                  },
                  "DATA_TRANSFER_INTERNAL_GBYTES": {
                    "type": "integer",
                    "example": 0
                  },
                  "DATA_TRANSFER_EXTERNAL_GBYTES": {
                    "type": "integer",
                    "example": 0
                  },
                  "PROXY_RESIDENTIAL_TRANSFER_GBYTES": {
                    "type": "integer",
                    "example": 0
                  },
                  "PROXY_SERPS": {
                    "type": "integer",
                    "example": 0
                  },
                  "PROXY_UNBLOCKER_UNITS": {
                    "type": "integer",
                    "example": 0
                  }
                }
              }
            }
          }
        }
      }
    }
  }
}