{
  "title": "PDF to Markdown - Text, Scanned OCR & PDF转Markdown",
  "type": "object",
  "schemaVersion": 1,
  "description": "Minimal input: {\"documentUrls\": [\"https://raw.githubusercontent.com/jsvine/pdfplumber/stable/examples/pdfs/background-checks.pdf\"], \"startPage\": 1, \"maxPagesPerDocument\": 5, \"ocrMode\": \"disabled\", \"ocrLanguage\": \"eng\"}. One paid row per nonempty PDF page.",
  "properties": {
    "documentUrls": {
      "title": "documentUrls",
      "type": "array",
      "description": "Public PDF URLs. Up to 100, 12 MB each. One billed nonempty page. Default digital text; enable ocrMode=auto for scanned pages without text. Use ocrLanguage=chi_sim or eng+chi_sim for Chinese.",
      "default": [
        "https://raw.githubusercontent.com/jsvine/pdfplumber/stable/examples/pdfs/background-checks.pdf"
      ],
      "editor": "stringList",
      "items": {
        "type": "string"
      },
      "maxItems": 100,
      "sectionCaption": "Targets / 目标"
    },
    "startPage": {
      "title": "startPage",
      "type": "integer",
      "description": "First PDF page to inspect, one-based, e.g. 2. Default 1. Pages before this are not inspected or billed.",
      "default": 1,
      "minimum": 1,
      "maximum": 10000
    },
    "maxPagesPerDocument": {
      "title": "maxPagesPerDocument",
      "type": "integer",
      "description": "Maximum PDF pages inspected per document from startPage, 1–20; default 5. This is a scan cap, not a guaranteed row count. Pages without extracted text consume it but have no result fee. A delivered text-bearing page is billed even if it contains no tables; OCR pages with recognized text use the same page price. maxResults separately caps paid rows.",
      "default": 5,
      "minimum": 1,
      "maximum": 20
    },
    "ocrMode": {
      "title": "ocrMode",
      "type": "string",
      "description": "\"disabled\" (default) extracts digital PDF text only. \"auto\" applies printed-text OCR only to pages with no extractable text. Does not OCR images on text-bearing pages or repair bad text layers. One billed nonempty page, same price. 扫描PDF识别。",
      "default": "disabled",
      "editor": "select",
      "enum": [
        "disabled",
        "auto"
      ]
    },
    "ocrLanguage": {
      "title": "ocrLanguage",
      "type": "string",
      "description": "\"eng\" (default), \"chi_sim\" for simplified Chinese, or \"eng+chi_sim\" for mixed printed text. Used only when ocrMode=auto needs OCR; not automatic language detection. 中文扫描件识别。",
      "default": "eng",
      "editor": "select",
      "enum": [
        "eng",
        "chi_sim",
        "eng+chi_sim"
      ]
    },
    "maxResults": {
      "title": "maxResults",
      "type": "integer",
      "description": "Maximum paid rows across the run, 1–1000, default 20. Example: 3 for a sample. Exclusions and empty results apply before the cap.",
      "default": 20,
      "minimum": 1,
      "maximum": 1000,
      "sectionCaption": "Limits / 数量限制"
    },
    "excludeIds": {
      "title": "excludeIds",
      "type": "array",
      "description": "Prior output id values to skip before billing, up to 5,000. Empty [] disables. Suppresses that identity even if content changed.",
      "default": [],
      "editor": "stringList",
      "items": {
        "type": "string"
      },
      "maxItems": 5000,
      "sectionCaption": "Repeat runs / 去重"
    },
    "excludeContentHashes": {
      "title": "excludeContentHashes",
      "type": "array",
      "description": "Prior output content_hash values (64 lowercase hex characters), up to 5,000. Empty [] disables. Matching snapshots are skipped before billing.",
      "default": [],
      "editor": "stringList",
      "items": {
        "type": "string"
      },
      "maxItems": 5000
    }
  },
  "additionalProperties": false
}
