[中文]

用途:

• 传 gh_username，返回公众号历史发文（文章 tab）的单页列表。
• 手动翻页：首页 offset 留空 → 从响应取 next_offset（base64 游标）→ 下次请求放进 offset 取下一页；is_end 为真即末页。
• 价格：0.01$/次
• ⏱️ 由于微信服务器原因，本接口响应较慢，请将客户端请求超时（timeout）设置为 30 秒；timeout 设置过小会造成已扣费但收不到响应的情况。

参数:

• username: 公众号 gh_username（gh_…）。示例 gh_363b924965e9
• page_size: 可选，默认 20。⚠️ 微信当前忽略该参数 —— 返回条数由公众号自身决定，实测取 1 与取 40 结果完全相同。翻页请用 offset / 响应里的 next_offset。
• offset: 可选，翻页游标（base64），首页留空；翻页时传上一页响应的 next_offset。示例 CAMQChiS5KfRBiAKOJLkp9EGQABIAVgAYABwAQ==
• item_show_type: 可选，内容栏目（对应公众号主页的「文章 / 视频 / 音频」分栏）。留空 / 0=文章（默认）、5=视频、7=音频、8=贴图。默认不传=文章，行为不变。
• raw: 可选，默认 True。True=原始；False=精简解析。

返回:

• 公众号文章列表（单页）及翻页游标

提示：offset 是 base64 游标（含 +/=），统一走 POST body。本接口只取单页、不自动翻页。

响应结构与 JSON Path:

raw=false（精简解析，snake_case）:

• 公众号 username: $.data.biz_username
• 是否末页（0/1）: $.data.is_end
• 本页返回文章数: $.data.count
• 下一页游标（翻页回传 offset；末页为 null）: $.data.next_offset
• 文章列表: $.data.articles[]
• 单篇（N 为下标）:
  • 文章 appmsgid: $.data.articles[N].app_msg_id
  • 标题: $.data.articles[N].title
  • 摘要: $.data.articles[N].digest
  • 链接: $.data.articles[N].url
  • 封面: $.data.articles[N].cover（多比例 $.data.articles[N].covers）
  • 发布 / 更新时间戳: $.data.articles[N].create_time / .update_time
  • 群发内位置 / 类型: $.data.articles[N].idx / .msg_type / .item_show_type
  • 图文数 / 付费标志: $.data.articles[N].pic_count / .is_paid / .is_pay_subscribe

raw=true（原始，本接口默认）:

• data 顶层字段与精简版一致（biz_username / is_end / count / next_offset），但 articles[] 为完整原始群发条目（camelCase 嵌套）:
  • 单条: $.data.articles[N].appMsg（含 baseInfo / detailInfo）、$.data.articles[N].baseInfo（msgId / msgType / dateTime / status）
  • 图文正文在 $.data.articles[N].appMsg.detailInfo（一次群发可含多篇：头条 / 次条）
  • 翻页同样用顶层 next_offset 回传 offset。做列表展示建议直接用 raw=false，路径更干净。

[English]

Purpose:

• Pass a gh_username to get a single page of the account's historical posts (article tab).
• Manual paging: leave offset empty for the first page → take next_offset (base64 cursor) from the response → pass it as offset in the next request; is_end truthy means the last page.
• Price: $0.01 per request
• ⏱️ Due to WeChat server latency, this endpoint responds slowly; please set your client request timeout to 30 seconds — a timeout that is too small may result in being billed without receiving the response.

Parameters:

• username: Official account gh_username (gh_…). E.g. gh_363b924965e9
• page_size: Optional, default 20. NOTE: WeChat currently ignores this parameter — the count is decided by the account itself (1 and 40 return identically). Paginate with offset / next_offset.
• offset: Optional pagination cursor (base64), leave empty for the first page; for the next page pass next_offset from the previous response. E.g. CAMQChiS5KfRBiAKOJLkp9EGQABIAVgAYABwAQ==
• item_show_type: Optional content tab (matches the "Articles / Videos / Audios" tabs on the account homepage). Empty / 0=articles (default), 5=videos, 7=audios, 8=image-text posts. Omitting it = articles, behavior unchanged.
• raw: Optional, default True. True=raw; False=simplified parsing.

Return:

• Single page of articles with pagination cursor

Tip: offset is a base64 cursor (contains +/=), hence POST body. This endpoint fetches a single page only and does not auto-paginate.

Response structure & JSON Path:

raw=false (simplified, snake_case):

• Account username: $.data.biz_username
• Is last page (0/1): $.data.is_end
• Articles returned on this page: $.data.count
• Next-page cursor (pass back as offset; null at the last page): $.data.next_offset
• Article list: $.data.articles[]
• Single article (N is the index):
  • Article appmsgid: $.data.articles[N].app_msg_id
  • Title: $.data.articles[N].title
  • Digest: $.data.articles[N].digest
  • URL: $.data.articles[N].url
  • Cover: $.data.articles[N].cover (multi-ratio $.data.articles[N].covers)
  • Publish / update timestamps: $.data.articles[N].create_time / .update_time
  • Position / type within the batch: $.data.articles[N].idx / .msg_type / .item_show_type
  • Image count / paid flags: $.data.articles[N].pic_count / .is_paid / .is_pay_subscribe

raw=true (raw, default for this endpoint):

• data top-level fields match the simplified version (biz_username / is_end / count / next_offset), but articles[] are full raw batch entries (nested camelCase):
  • Per entry: $.data.articles[N].appMsg (with baseInfo / detailInfo), $.data.articles[N].baseInfo (msgId / msgType / dateTime / status)
  • Article bodies are in $.data.articles[N].appMsg.detailInfo (one batch may contain multiple articles: headline / secondary)
  • Pagination also uses top-level next_offset passed back as offset. For list display, prefer raw=false for cleaner paths.

{
  "username": "gh_363b924965e9"
}