{
  "id": "web.extract",
  "object": "tool",
  "name": "Extract from a page",
  "summary": "Structured data from any page.",
  "description": "Renders a page, reads it as Markdown, and has an LLM fill in a JSON Schema you give (and/or follow a plain-language instruction). The result is validated against the schema, with one automatic retry.",
  "category": "web",
  "hosted_by": "Opentools",
  "runtime": "browser",
  "side_effects": false,
  "status": "live",
  "credits": 3,
  "price": "$0.0030",
  "providers": [
    {
      "id": "opentools",
      "name": "Opentools",
      "price": "$0.0030"
    }
  ],
  "mcp_name": "web_extract",
  "input_schema": {
    "type": "object",
    "properties": {
      "url": {
        "type": "string",
        "format": "uri",
        "description": "Public http(s) URL of the page"
      },
      "schema": {
        "type": "object",
        "propertyNames": {
          "type": "string"
        },
        "additionalProperties": {},
        "description": "A JSON Schema for the result, e.g. { \"type\": \"object\", \"properties\": { \"price\": { \"type\": \"number\" } } }"
      },
      "instruction": {
        "description": "What to extract, e.g. \"all job openings with location\"",
        "type": "string",
        "maxLength": 4000
      },
      "render": {
        "default": true,
        "description": "Render with a browser. Off: plain fetch, faster for static pages.",
        "type": "boolean"
      },
      "wait_until": {
        "default": "networkidle",
        "description": "When the page counts as loaded. networkidle waits until the network is quiet.",
        "type": "string",
        "enum": [
          "load",
          "domcontentloaded",
          "networkidle"
        ]
      },
      "wait_for_selector": {
        "description": "Also wait until this CSS selector appears",
        "type": "string",
        "maxLength": 500
      },
      "cookies": {
        "description": "Cookies to set before loading, e.g. a consent or session cookie",
        "maxItems": 50,
        "type": "array",
        "items": {
          "type": "object",
          "properties": {
            "name": {
              "type": "string",
              "description": "Cookie name"
            },
            "value": {
              "type": "string",
              "description": "Cookie value"
            },
            "domain": {
              "description": "Cookie domain. The page's domain if omitted.",
              "type": "string"
            },
            "path": {
              "description": "Cookie path. / if omitted.",
              "type": "string"
            }
          },
          "required": [
            "name",
            "value"
          ]
        }
      },
      "headers": {
        "description": "Extra HTTP headers sent with every request",
        "type": "object",
        "propertyNames": {
          "type": "string"
        },
        "additionalProperties": {
          "type": "string"
        }
      }
    },
    "required": [
      "url"
    ]
  },
  "kind": "tool"
}