{
  "id": "parse-web",
  "name": "parse-web",
  "kind": "step",
  "contentType": "component",
  "description": "Extract source-attributed main content from a fetched web page.",
  "value": "Extract source-attributed main content from a fetched web page.",
  "input": {
    "html": {
      "type": "string",
      "required": true,
      "notes": "Fetched HTML bytes."
    },
    "sourceUrl": {
      "type": "string",
      "required": true,
      "notes": "Original source URL."
    }
  },
  "output": {
    "document": {
      "type": "object",
      "notes": "Title, main text, canonical URL and extraction diagnostics."
    }
  },
  "needsEnv": [],
  "implementation": {
    "status": "planned",
    "source": null,
    "limitation": "Fetching HTML is not parsing it. Reject unsupported pages or report unknown; do not fabricate an extraction."
  },
  "acceptance": {
    "status": "pending",
    "evidence": []
  },
  "usedBy": []
}