Agent/examples/process_pipeline/prompts/temp_schema/source.schema.json

{
  "$schema": "http://json-schema.org/draft-07/schema#",
  "title": "Source (Normalized)",
  "description": "标准化后的 source 数据格式，与 sota_research 的 result.json 对齐",
  "type": "object",
  "required": ["total-ref", "sources-boundary"],
  "properties": {
    "total-ref": {
      "type": "integer",
      "description": "总记录数",
      "minimum": 0
    },
    "cache_dir": {
      "type": "string",
      "description": "Cache 目录路径"
    },
    "sources-boundary": {
      "type": "array",
      "description": "标准化后的 source 记录列表",
      "items": {
        "type": "object",
        "required": ["index-ref", "title-ref", "body", "images-boundary", "url-ref", "_raw-boundary"],
        "properties": {
          "index-ref": {
            "type": "integer",
            "description": "记录序号",
            "minimum": 1
          },
          "category": {
            "type": "string",
            "description": "分类标签"
          },
          "user_kept-ref": {
            "type": "boolean",
            "description": "用户是否保留"
          },
          "user_comment": {
            "type": "string",
            "description": "用户备注"
          },
          "description": {
            "type": "string",
            "description": "描述"
          },
          "method": {
            "type": "string",
            "description": "方法"
          },
          "cover": {
            "type": "string",
            "description": "封面图 CDN URL"
          },
          "title-ref": {
            "type": "string",
            "description": "标题"
          },
          "author": {
            "type": "string",
            "description": "作者"
          },
          "body": {
            "type": "string",
            "description": "正文内容"
          },
          "images-boundary": {
            "type": "array",
            "description": "图片 CDN URL 列表",
            "items": {
              "type": "string"
            }
          },
          "url-ref": {
            "type": "string",
            "description": "原始帖子链接"
          },
          "note": {
            "type": "string",
            "description": "备注信息（platform, likes, comments）"
          },
          "_raw-boundary": {
            "type": "object",
            "description": "原始标识信息（稳定引用区）",
            "required": ["case_id-ref", "platform-ref", "channel_content_id-ref"],
            "properties": {
              "case_id-ref": {
                "type": "string",
                "description": "统一格式的 case ID: {platform}_{content_id}",
                "pattern": "^[a-z]+_[A-Za-z0-9_-]+$"
              },
              "platform-ref": {
                "type": "string",
                "description": "平台标识",
                "minLength": 1
              },
              "channel_content_id-ref": {
                "type": "string",
                "description": "平台内容 ID",
                "minLength": 1
              }
            }
          }
        }
      }
    }
  }
}