@openschool_01/skills 0.1.3 → 0.1.4

package/bundled-skills/wechat-assistant/skill/OPENSCHOOL-META.md

@@ -2,18 +2,19 @@
 
 - 收录名称：公众号助手
 - 收录方式：OpenSchool 严选整合
-- 当前版本：纯内容版
-
-当前版本整合了公众号写作流程里最常用的三块能力：
+- 当前版本：内容与草稿箱双模版
 
+本版本重点能力：
 1. 标题方向整理
-2. 正文写作约束
-3. 微信 HTML 渲染与质量检查
-
-本版本刻意不包含：
+2. 正文写作与质量检查
+3. 公众号预览 HTML 渲染（默认无尾注）
+4. 可选上传草稿箱（用户显式提供 appid/appsecret 时执行）
 
+默认不包含：
 - 生图
-- 封面生成
-- 微信 API 发布
+- 封面自动生成
+- 自动正式发布
 
-这样可以先保证“写得出、改得动、落得下、能预览”这条链路稳定。
+上线约束：
+- 不内置任何个人信息尾注
+- 不在仓库存储真实公众号凭据

package/bundled-skills/wechat-assistant/skill/SKILL.md

@@ -2,108 +2,112 @@
 
 适用于把一个选题整理成可发布到微信公众号的完整内容稿。
 
-当前版本专注 4 件事：
-
-1. 明确选题和读者视角
-2. 生成更像公众号风格的标题方向
-3. 写出 Markdown 正文
-4. 渲染成微信公众号可用的 HTML
-
-这不是一个“全自动发布器”。当前版本不负责：
-
+当前版本重点做 5 件事：
+1. 选题和标题方向整理
+2. 正文 Markdown 写作
+3. 微信预览 HTML 渲染
+4. 文章质量检查
+5. 可选上传到公众号草稿箱（需要用户主动提供凭据）
+
+默认不做：
 - 生图
-- 封面生成
-- 微信草稿箱发布
+- 封面图自动生成
+- 自动正式发布
 
-## 什么时候用
+默认不追加任何个人尾注信息。
 
-当用户要做这些事时使用：
+## 什么时候用
 
 - 写公众号文章
-- 把直播内容整理成公众号稿
-- 生成公众号标题
-- 输出微信公众号排版版式的 HTML
-- 检查文章是否适合公众号发布
+- 把直播内容整理成公众号稿件
+- 生成公众号标题候选
+- 输出可预览、可复制到后台的 HTML
+- 需要通过 API 上传到草稿箱
 
-## 推荐工作流
+## 推荐流程
 
-### 1. 先收集输入
+### 1. 收集输入
 
 至少确认这 4 个信息：
-
 - 文章主题
 - 目标读者
-- 读者当前最常见的问题
+- 读者当前最常见问题
 - 这篇文章承诺解决什么
 
-### 2. 先给标题方向
-
-优先输出 3 到 5 个标题，并解释它们分别适合：
-
-- 偏方法拆解
-- 偏观点判断
-- 偏案例复盘
+### 2. 先出标题方向
 
-标题规则见：
+输出 3 到 5 个标题，并说明适用场景。
 
-- `references/title-guidelines.md`
+### 3. 再写正文 Markdown
 
-### 3. 再写正文
+正文建议：
+- 4 到 6 个二级标题
+- 先给判断，再解释
+- 至少 1 个真实场景
+- 至少 1 个常见误区
+- 至少 1 组可执行动作
 
-正文默认要求：
+### 4. 渲染公众号 HTML（无尾注）
 
-- 使用 4 到 6 个二级标题
-- 先给判断，再展开解释
-- 至少包含 1 个真实场景
-- 至少包含 1 个常见误区
-- 至少包含 1 组可执行动作
-- 结尾自然引向下一篇或下一步动作
+```bash
+python scripts/render_wechat_html.py --input "<article.md>" --output "<article>.html"
+```
 
-写作约束见：
+### 5. 质量检查（可选）
 
-- `references/writing-rules.md`
+```bash
+python scripts/audit_article_quality.py --input "<article.md>"
+```
 
-### 4. 渲染 HTML
+### 6. 上传草稿箱（可选）
 
-正文 Markdown 完成后，运行：
+当用户明确需要上传草稿箱时，使用：
 
 ```bash
-python scripts/render_wechat_html.py --input "<article.md>" --output "<article>.html"
+python scripts/wechat_html_api_publish.py \
+  --markdown "<article.md>" \
+  --config "config/wechat-publish-api.example.json" \
+  --dry-run
 ```
 
-### 5. 做一次质量检查
+确认 dry-run 输出没问题后，再去掉 `--dry-run`。
 
-如果用户在意发布质量，再运行：
+也可以直接传入账号参数：
 
 ```bash
-python scripts/audit_article_quality.py --input "<article.md>"
+python scripts/wechat_html_api_publish.py \
+  --markdown "<article.md>" \
+  --appid "<appid>" \
+  --appsecret "<appsecret>" \
+  --author "<author>"
 ```
 
+## 安全约束
+
+- 不要把用户的 appid/appsecret 写死进脚本。
+- 不要把真实凭据提交到仓库。
+- 默认使用示例配置文件，用户自行填写本地私密配置。
+
 ## 目录说明
 
-- `references/title-guidelines.md`
-  标题风格和筛选规则
-- `references/writing-rules.md`
-  正文写作规则和结构要求
-- `assets/wechat-article-template.html`
-  微信 HTML 模板
-- `config/wechat-style-config.json`
-  样式配置
-- `scripts/render_wechat_html.py`
-  Markdown 转微信 HTML
-- `scripts/audit_article_quality.py`
-  质量检查脚本
+- `references/title-guidelines.md`：标题参考
+- `references/writing-rules.md`：正文写作规则
+- `assets/wechat-article-template.html`：公众号预览模板
+- `config/wechat-style-config.json`：样式配置
+- `config/wechat-publish-api.example.json`：草稿箱上传配置示例
+- `scripts/render_wechat_html.py`：Markdown 转公众号预览 HTML（无尾注）
+- `scripts/render_wechat_api_content.py`：Markdown 转 API 友好内容片段
+- `scripts/wechat_html_api_publish.py`：调用公众号接口创建草稿
+- `scripts/audit_article_quality.py`：质量检查
 
 ## 交付要求
 
 默认交付：
-
-- 3 到 5 个标题备选
+- 3 到 5 个标题候选
 - 1 篇 Markdown 正文
-- 1 个 HTML 文件
-
-如果脚本已经执行，还要明确告诉用户：
+- 1 个 HTML 预览文件
 
-- Markdown 文件路径
-- HTML 文件路径
-- 质量检查结果
+如果执行了草稿箱上传，还要明确返回：
+- 是否 dry-run
+- 草稿接口返回结果
+- 产物文件路径

package/bundled-skills/wechat-assistant/skill/config/wechat-publish-api.example.json

@@ -0,0 +1,6 @@
+{
+  "appid": "YOUR_WECHAT_APPID",
+  "appsecret": "YOUR_WECHAT_APPSECRET",
+  "author": "OpenSchool",
+  "content_source_url": ""
+}

package/bundled-skills/wechat-assistant/skill/scripts/render_wechat_api_content.py

@@ -0,0 +1,191 @@
+#!/usr/bin/env python3
+"""Render markdown into WeChat Official Account API-friendly HTML fragment."""
+
+from __future__ import annotations
+
+import argparse
+import html
+import re
+from pathlib import Path
+
+WECHAT_INLINE = {
+  "p": "margin:0 0 18px;line-height:1.85em;font-size:16px;color:#222;text-align:justify;word-break:break-word;",
+  "h2": "margin:30px 0 14px;line-height:1.75em;font-size:17px;font-weight:700;color:#0f6fff;",
+  "h3": "margin:22px 0 12px;line-height:1.75em;font-size:15px;font-weight:700;color:#0f6fff;",
+  "li": "margin:0 0 10px;line-height:1.85em;font-size:16px;color:#222;",
+  "blockquote": "margin:0 0 18px;padding:10px 14px;background:#f7faff;border-left:3px solid #0f6fff;color:#355070;line-height:1.85em;font-size:15px;",
+  "pre": "margin:0 0 18px;padding:12px 14px;background:#f7f8fa;border:0;border-left:3px solid rgba(15,111,255,.45);overflow-x:auto;",
+  "code": "font-size:14px;line-height:1.75em;color:#333;font-family:Consolas,Menlo,monospace;white-space:pre-wrap;",
+  "sep": "margin:22px 0 20px;text-align:center;color:#bbb;letter-spacing:1px;",
+  "cta": "margin-top:28px;padding-top:14px;border-top:1px solid #e6e6e6;",
+}
+
+
+def convert_inline(text: str) -> str:
+  text = html.escape(text)
+  text = re.sub(r"`([^`]+)`", r"<code style=\"%s\">\1</code>" % WECHAT_INLINE["code"], text)
+  text = re.sub(r"\*\*([^*]+)\*\*", r"<strong style=\"color:#0f6fff;font-weight:700;\">\1</strong>", text)
+  text = re.sub(r"\*([^*]+)\*", r"<em>\1</em>", text)
+  text = re.sub(r"\[([^\]]+)\]\(([^)]+)\)", r'<a href="\2">\1</a>', text)
+  return text
+
+
+def markdown_to_api_blocks(content: str):
+  lines = content.splitlines()
+  title = ""
+  blocks = []
+  in_code = False
+  code = []
+  para = []
+  quote = []
+  list_items: list[str] = []
+  list_kind = ""
+  has_cta = False
+
+  def flush_para():
+    nonlocal para, has_cta
+    if not para:
+      return
+    txt = re.sub(r"\s+", " ", " ".join(p.strip() for p in para if p.strip())).strip()
+    para = []
+    if not txt:
+      return
+    if any(k in txt for k in ["留言", "评论", "关键词", "下一篇", "下篇", "领取"]):
+      has_cta = True
+    blocks.append(f'<p style="{WECHAT_INLINE["p"]}">{convert_inline(txt)}</p>')
+
+  def flush_quote():
+    nonlocal quote
+    if not quote:
+      return
+    txt = re.sub(r"\s+", " ", " ".join(q.strip() for q in quote if q.strip())).strip()
+    quote = []
+    if not txt:
+      return
+    blocks.append(f'<blockquote style="{WECHAT_INLINE["blockquote"]}">{convert_inline(txt)}</blockquote>')
+
+  def flush_list():
+    nonlocal list_items, list_kind
+    if not list_items:
+      return
+    tag = "ol" if list_kind == "ol" else "ul"
+    items = "".join(f'<li style="{WECHAT_INLINE["li"]}">{convert_inline(item)}</li>' for item in list_items)
+    blocks.append(f"<{tag}>{items}</{tag}>")
+    list_items = []
+    list_kind = ""
+
+  for raw in lines:
+    s = raw.rstrip()
+    st = s.strip()
+    if st.startswith("```"):
+      flush_para()
+      flush_quote()
+      flush_list()
+      if in_code:
+        code_html = html.escape("\n".join(code))
+        blocks.append(f'<pre style="{WECHAT_INLINE["pre"]}"><code style="{WECHAT_INLINE["code"]}">{code_html}</code></pre>')
+        code = []
+        in_code = False
+      else:
+        in_code = True
+      continue
+
+    if in_code:
+      code.append(s)
+      continue
+
+    if not st:
+      flush_para()
+      flush_quote()
+      flush_list()
+      continue
+
+    if not title and st.startswith("# "):
+      flush_para()
+      flush_quote()
+      flush_list()
+      title = st[2:].strip()
+      continue
+
+    if st.startswith("## "):
+      flush_para()
+      flush_quote()
+      flush_list()
+      blocks.append(f'<h2 style="{WECHAT_INLINE["h2"]}">{convert_inline(st[3:])}</h2>')
+      continue
+
+    if st.startswith("### "):
+      flush_para()
+      flush_quote()
+      flush_list()
+      blocks.append(f'<h3 style="{WECHAT_INLINE["h3"]}">{convert_inline(st[4:])}</h3>')
+      continue
+
+    if st.startswith("> "):
+      flush_para()
+      flush_list()
+      quote.append(st[2:])
+      continue
+
+    if re.match(r"^\d+\.\s+", st):
+      flush_para()
+      flush_quote()
+      kind = "ol"
+      if list_kind and list_kind != kind:
+        flush_list()
+      list_kind = kind
+      list_items.append(re.sub(r"^\d+\.\s+", "", st))
+      continue
+
+    if re.match(r"^[-*]\s+", st):
+      flush_para()
+      flush_quote()
+      kind = "ul"
+      if list_kind and list_kind != kind:
+        flush_list()
+      list_kind = kind
+      list_items.append(re.sub(r"^[-*]\s+", "", st))
+      continue
+
+    if st == "---":
+      flush_para()
+      flush_quote()
+      flush_list()
+      blocks.append(f'<p style="{WECHAT_INLINE["sep"]}">———</p>')
+      continue
+
+    para.append(st)
+
+  flush_para()
+  flush_quote()
+  flush_list()
+  return title, "\n".join(blocks), has_cta
+
+
+def render_api_content(markdown: str, title_override: str = "") -> tuple[str, str, bool]:
+  title, body, has_cta = markdown_to_api_blocks(markdown)
+  title = title_override or title or "公众号文章"
+  cta = (
+    ""
+    if has_cta
+    else f'<section style="{WECHAT_INLINE["cta"]}"><p style="{WECHAT_INLINE["p"]}color:#355070;">如果这篇内容对你有帮助，欢迎留言关键词，我会继续补充相关资料。</p></section>'
+  )
+  return title, body + cta, has_cta
+
+
+def main() -> None:
+  parser = argparse.ArgumentParser()
+  parser.add_argument("--input", required=True)
+  parser.add_argument("--output", required=True)
+  parser.add_argument("--title", default="")
+  args = parser.parse_args()
+
+  markdown = Path(args.input).read_text(encoding="utf-8")
+  _, html_doc, _ = render_api_content(markdown, args.title)
+  out = Path(args.output)
+  out.parent.mkdir(parents=True, exist_ok=True)
+  out.write_text(html_doc, encoding="utf-8")
+
+
+if __name__ == "__main__":
+  main()

package/bundled-skills/wechat-assistant/skill/scripts/render_wechat_html.py

@@ -1,5 +1,8 @@
 #!/usr/bin/env python3
-"""Render markdown into WeChat-friendly article HTML."""
+"""Render markdown into WeChat-friendly preview HTML.
+
+This version intentionally does NOT append personal footer/tail notes.
+"""
 
 from __future__ import annotations
 
@@ -9,163 +12,224 @@ import json
 import re
 from pathlib import Path
 
-
 DEFAULT_STYLE = """
-body { margin: 0; background: #ffffff; color: #111111; font-family: "PingFang SC", "Microsoft YaHei", sans-serif; }
-.wx-wrap { max-width: 720px; margin: 0 auto; padding: 24px 18px 48px; box-sizing: border-box; }
-.wx-title { margin: 0 0 12px; font-size: 28px; line-height: 1.35; font-weight: 700; color: #111111; }
-.wx-subtitle { margin: 0 0 20px; font-size: 15px; line-height: 1.8; color: #666666; }
-.wx-content p, .wx-content li { margin: 0 0 14px; font-size: 16px; line-height: 1.8; color: #111111; text-align: justify; text-justify: inter-ideograph; word-break: break-word; }
-.wx-content h2, .wx-content h3 { margin: 26px 0 10px; font-size: 20px; line-height: 1.6; font-weight: 700; color: rgb(0,128,255); }
-.wx-content strong { color: rgb(0,128,255); font-weight: 700; }
-.wx-content ul, .wx-content ol { margin: 0 0 14px 1.4em; padding: 0; }
-.wx-content pre { margin: 0 0 18px; padding: 14px 16px; background: #f7f8fa; border-left: 3px solid rgba(0,128,255,.45); overflow-x: auto; border-radius: 6px; }
-.wx-content code { font-size: 14px; line-height: 1.7; color: #333333; font-family: "Consolas", "Menlo", monospace; white-space: pre-wrap; }
-.wx-sep { margin: 12px 0 16px; font-size: 16px; line-height: 1.8; color: #999999; text-align: center; }
-.wx-cta { margin-top: 28px; padding-top: 16px; border-top: 1px solid #ececec; }
-.wx-cta p { margin: 0; font-size: 15px; line-height: 1.8; color: #444444; }
-.wx-footer { margin-top: 44px; padding-top: 18px; border-top: 1px solid #f0f0f0; }
-.wx-footer p { margin: 0; font-size: 13px; line-height: 1.9; color: #a0a0a0; text-align: center; }
+body { margin: 0; background: #ffffff; color: #1c2430; font-family: "PingFang SC", "Microsoft YaHei", sans-serif; }
+.wx-wrap { max-width: 720px; margin: 0 auto; padding: 26px 18px 42px; box-sizing: border-box; }
+.wx-title { margin: 0 0 10px; font-size: 30px; line-height: 1.38; font-weight: 700; color: #0f172a; letter-spacing: .2px; }
+.wx-subtitle { margin: 0 0 18px; font-size: 14px; line-height: 1.8; color: #6b7280; }
+.wx-content p, .wx-content li { margin: 0 0 14px; font-size: 16px; line-height: 1.9; color: #1f2937; text-align: justify; text-justify: inter-ideograph; word-break: break-word; }
+.wx-content h2 { margin: 28px 0 14px; padding: 10px 12px; border-left: 4px solid #0f6fff; border-radius: 6px; background: linear-gradient(90deg, rgba(15,111,255,.10), rgba(15,111,255,.02)); font-size: 20px; line-height: 1.6; font-weight: 700; color: #0f172a; }
+.wx-content h3 { margin: 22px 0 10px; font-size: 17px; line-height: 1.7; font-weight: 700; color: #0f6fff; }
+.wx-content strong { color: #0f172a; font-weight: 700; background-image: linear-gradient(transparent 62%, rgba(251,191,36,.35) 0); }
+.wx-content ul, .wx-content ol { margin: 2px 0 16px 1.35em; padding: 0; }
+.wx-content li { margin-bottom: 8px; }
+.wx-content blockquote { margin: 0 0 16px; padding: 12px 14px; border-left: 3px solid rgba(15,111,255,.45); border-radius: 6px; background: #f8fbff; color: #334155; }
+.wx-content pre { margin: 0 0 18px; padding: 13px 14px; background: #f8fafc; border: 1px solid #e2e8f0; border-radius: 8px; overflow-x: auto; }
+.wx-content code { font-size: 14px; line-height: 1.75; color: #334155; font-family: "Consolas", "Menlo", monospace; white-space: pre-wrap; }
+.wx-sep { margin: 16px 0 18px; font-size: 14px; line-height: 1.8; color: #9ca3af; text-align: center; letter-spacing: 1px; }
+.wx-cta { margin-top: 24px; padding: 13px 14px; border: 1px solid #e5e7eb; border-radius: 8px; background: #fafafa; }
+.wx-cta p { margin: 0; font-size: 15px; line-height: 1.85; color: #374151; }
 """
 
 
 def load_style(config_path: Path | None) -> str:
-    if not config_path or not config_path.exists():
-        return DEFAULT_STYLE
+  if not config_path or not config_path.exists():
+    return DEFAULT_STYLE
 
-    config = json.loads(config_path.read_text(encoding="utf-8"))
-    body_font_size = int(config.get("bodyFontSize", 16))
-    body_line_height = float(config.get("bodyLineHeight", 1.8))
-    content_width = int(config.get("contentWidth", 720))
-    accent = str(config.get("accentColor", "rgb(0,128,255)"))
-    text = str(config.get("textColor", "#111111"))
+  config = json.loads(config_path.read_text(encoding="utf-8"))
+  body_font_size = int(config.get("bodyFontSize", 16))
+  body_line_height = float(config.get("bodyLineHeight", 1.9))
+  content_width = int(config.get("contentWidth", 720))
+  accent = str(config.get("accentColor", "#0f6fff"))
+  text = str(config.get("textColor", "#1f2937"))
 
-    return DEFAULT_STYLE.replace("720px", f"{content_width}px").replace(
-        "rgb(0,128,255)", accent
-    ).replace("#111111", text).replace("font-size: 16px; line-height: 1.8;", f"font-size: {body_font_size}px; line-height: {body_line_height};")
+  style = DEFAULT_STYLE.replace("720px", f"{content_width}px")
+  style = style.replace("#0f6fff", accent)
+  style = style.replace("#1f2937", text)
+  style = style.replace("font-size: 16px; line-height: 1.9;", f"font-size: {body_font_size}px; line-height: {body_line_height};", 1)
+  return style
 
 
 def convert_inline(text: str) -> str:
-    text = re.sub(r"!\[\[[^\]]+\]\]", "", text)
-    text = html.escape(text)
-    text = re.sub(r"`([^`]+)`", r"<code>\1</code>", text)
-    text = re.sub(r"\*\*([^*]+)\*\*", r"<strong>\1</strong>", text)
-    text = re.sub(r"\*([^*]+)\*", r"<em>\1</em>", text)
-    text = re.sub(r"\[([^\]]+)\]\(([^)]+)\)", r'<a href="\2">\1</a>', text)
-    return text
+  text = re.sub(r"!\[\[[^\]]+\]\]", "", text)
+  text = html.escape(text)
+  text = re.sub(r"`([^`]+)`", r"<code>\1</code>", text)
+  text = re.sub(r"\*\*([^*]+)\*\*", r"<strong>\1</strong>", text)
+  text = re.sub(r"\*([^*]+)\*", r"<em>\1</em>", text)
+  text = re.sub(r"\[([^\]]+)\]\(([^)]+)\)", r'<a href="\2">\1</a>', text)
+  return text
 
 
 def markdown_to_blocks(content: str) -> tuple[str, str, bool]:
-    lines = content.splitlines()
-    title = ""
-    blocks: list[str] = []
-    in_code = False
-    code_lines: list[str] = []
-    paragraph_lines: list[str] = []
-    has_cta = False
-
-    def flush_paragraph() -> None:
-        nonlocal paragraph_lines, has_cta
-        if not paragraph_lines:
-            return
-        text = re.sub(r"\s+", " ", " ".join(line.strip() for line in paragraph_lines if line.strip())).strip()
-        paragraph_lines = []
-        if not text:
-            return
-        if any(keyword in text for keyword in ["留言", "评论", "下一篇", "下篇", "领取"]):
-            has_cta = True
-        blocks.append(f"<p>{convert_inline(text)}</p>")
-
-    for raw_line in lines:
-        line = raw_line.rstrip()
-        stripped = line.strip()
-
-        if stripped.startswith("```"):
-            flush_paragraph()
-            if in_code:
-                blocks.append("<pre><code>" + html.escape("\n".join(code_lines)) + "</code></pre>")
-                code_lines = []
-                in_code = False
-            else:
-                in_code = True
-            continue
-
-        if in_code:
-            code_lines.append(line)
-            continue
-
-        if not stripped:
-            flush_paragraph()
-            continue
-
-        if not title and stripped.startswith("# "):
-            title = stripped[2:].strip()
-            continue
-
-        if stripped.startswith("## "):
-            flush_paragraph()
-            blocks.append(f"<h2>{convert_inline(stripped[3:])}</h2>")
-            continue
-
-        if stripped.startswith("### "):
-            flush_paragraph()
-            blocks.append(f"<h3>{convert_inline(stripped[4:])}</h3>")
-            continue
-
-        if re.match(r"^\d+\.\s+", stripped):
-            flush_paragraph()
-            blocks.append("<ol><li>" + convert_inline(re.sub(r"^\d+\.\s+", "", stripped)) + "</li></ol>")
-            continue
-
-        if re.match(r"^[-*]\s+", stripped):
-            flush_paragraph()
-            blocks.append("<ul><li>" + convert_inline(re.sub(r"^[-*]\s+", "", stripped)) + "</li></ul>")
-            continue
-
-        if stripped == "---":
-            flush_paragraph()
-            blocks.append('<p class="wx-sep">···</p>')
-            continue
-
-        paragraph_lines.append(stripped)
-
-    flush_paragraph()
-    return title, "\n".join(blocks), has_cta
+  lines = content.splitlines()
+  title = ""
+  blocks: list[str] = []
+  in_code = False
+  code_lines: list[str] = []
+  paragraph_lines: list[str] = []
+  quote_lines: list[str] = []
+  list_items: list[str] = []
+  list_kind = ""
+  has_cta = False
+
+  def flush_paragraph() -> None:
+    nonlocal paragraph_lines, has_cta
+    if not paragraph_lines:
+      return
+    text = re.sub(r"\s+", " ", " ".join(line.strip() for line in paragraph_lines if line.strip())).strip()
+    paragraph_lines = []
+    if not text:
+      return
+    if any(keyword in text for keyword in ["留言", "评论", "下一篇", "下篇", "领取"]):
+      has_cta = True
+    blocks.append(f"<p>{convert_inline(text)}</p>")
+
+  def flush_quote() -> None:
+    nonlocal quote_lines
+    if not quote_lines:
+      return
+    text = re.sub(r"\s+", " ", " ".join(line.strip() for line in quote_lines if line.strip())).strip()
+    quote_lines = []
+    if text:
+      blocks.append(f"<blockquote>{convert_inline(text)}</blockquote>")
+
+  def flush_list() -> None:
+    nonlocal list_items, list_kind
+    if not list_items:
+      return
+    tag = "ol" if list_kind == "ol" else "ul"
+    items = "".join(f"<li>{convert_inline(item)}</li>" for item in list_items)
+    blocks.append(f"<{tag}>{items}</{tag}>")
+    list_items = []
+    list_kind = ""
+
+  for raw_line in lines:
+    line = raw_line.rstrip()
+    stripped = line.strip()
+
+    if stripped.startswith("```"):
+      flush_paragraph()
+      flush_quote()
+      flush_list()
+      if in_code:
+        blocks.append("<pre><code>" + html.escape("\n".join(code_lines)) + "</code></pre>")
+        code_lines = []
+        in_code = False
+      else:
+        in_code = True
+      continue
+
+    if in_code:
+      code_lines.append(line)
+      continue
+
+    if not stripped:
+      flush_paragraph()
+      flush_quote()
+      flush_list()
+      continue
+
+    if not title and stripped.startswith("# "):
+      flush_paragraph()
+      flush_quote()
+      flush_list()
+      title = stripped[2:].strip()
+      continue
+
+    if stripped.startswith("## "):
+      flush_paragraph()
+      flush_quote()
+      flush_list()
+      blocks.append(f"<h2>{convert_inline(stripped[3:])}</h2>")
+      continue
+
+    if stripped.startswith("### "):
+      flush_paragraph()
+      flush_quote()
+      flush_list()
+      blocks.append(f"<h3>{convert_inline(stripped[4:])}</h3>")
+      continue
+
+    if stripped.startswith("> "):
+      flush_paragraph()
+      flush_list()
+      quote_lines.append(stripped[2:])
+      continue
+
+    if re.match(r"^\d+\.\s+", stripped):
+      flush_paragraph()
+      flush_quote()
+      kind = "ol"
+      if list_kind and list_kind != kind:
+        flush_list()
+      list_kind = kind
+      list_items.append(re.sub(r"^\d+\.\s+", "", stripped))
+      continue
+
+    if re.match(r"^[-*]\s+", stripped):
+      flush_paragraph()
+      flush_quote()
+      kind = "ul"
+      if list_kind and list_kind != kind:
+        flush_list()
+      list_kind = kind
+      list_items.append(re.sub(r"^[-*]\s+", "", stripped))
+      continue
+
+    if stripped == "---":
+      flush_paragraph()
+      flush_quote()
+      flush_list()
+      blocks.append('<p class="wx-sep">———</p>')
+      continue
+
+    paragraph_lines.append(stripped)
+
+  flush_paragraph()
+  flush_quote()
+  flush_list()
+  return title, "\n".join(blocks), has_cta
 
 
 def render_html(markdown: str, title_override: str, style_config_path: Path | None) -> str:
-    title, body, has_cta = markdown_to_blocks(markdown)
-    title = title_override or title or "公众号文章"
-    template_path = Path(__file__).resolve().parent.parent / "assets" / "wechat-article-template.html"
-    template = template_path.read_text(encoding="utf-8")
-    cta = "" if has_cta else '<footer class="wx-cta"><p>如果这篇内容对你有帮助，下一步可以继续把同主题素材整理成系列文章。</p></footer>'
-    footer = '<footer class="wx-footer"><p>由 OpenSchool 严选技能整理输出</p></footer>'
-    subtitle_block = ""
-
-    output = template.replace("{{TITLE}}", html.escape(title))
-    output = output.replace("{{SUBTITLE_BLOCK}}", subtitle_block)
-    output = output.replace("{{STYLE}}", load_style(style_config_path).strip())
-    output = output.replace("{{CONTENT}}", body)
-    output = output.replace("{{CTA_BLOCK}}", cta)
-    output = output.replace("{{FOOTER_BLOCK}}", footer)
-    return output
+  title, body, has_cta = markdown_to_blocks(markdown)
+  title = title_override or title or "公众号文章"
+  template_path = Path(__file__).resolve().parent.parent / "assets" / "wechat-article-template.html"
+  template = template_path.read_text(encoding="utf-8")
+  cta = (
+    ""
+    if has_cta
+    else '<footer class="wx-cta"><p>如果这篇内容对你有帮助，欢迎留言关键词，我会继续补充同主题资料。</p></footer>'
+  )
+
+  output = template.replace("{{TITLE}}", html.escape(title))
+  output = output.replace("{{SUBTITLE_BLOCK}}", "")
+  output = output.replace("{{STYLE}}", load_style(style_config_path).strip())
+  output = output.replace("{{CONTENT}}", body)
+  output = output.replace("{{CTA_BLOCK}}", cta)
+  output = output.replace("{{FOOTER_BLOCK}}", "")
+  return output
 
 
 def main() -> None:
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--input", required=True)
-    parser.add_argument("--output", required=True)
-    parser.add_argument("--title", default="")
-    parser.add_argument("--style-config", default="")
-    args = parser.parse_args()
-
-    style_config_path = Path(args.style_config) if args.style_config else Path(__file__).resolve().parent.parent / "config" / "wechat-style-config.json"
-    markdown = Path(args.input).read_text(encoding="utf-8")
-    output = render_html(markdown, args.title, style_config_path)
-    output_path = Path(args.output)
-    output_path.parent.mkdir(parents=True, exist_ok=True)
-    output_path.write_text(output, encoding="utf-8")
+  parser = argparse.ArgumentParser()
+  parser.add_argument("--input", required=True)
+  parser.add_argument("--output", required=True)
+  parser.add_argument("--title", default="")
+  parser.add_argument("--style-config", default="")
+  args = parser.parse_args()
+
+  style_config_path = (
+    Path(args.style_config)
+    if args.style_config
+    else Path(__file__).resolve().parent.parent / "config" / "wechat-style-config.json"
+  )
+  markdown = Path(args.input).read_text(encoding="utf-8")
+  output = render_html(markdown, args.title, style_config_path)
+  output_path = Path(args.output)
+  output_path.parent.mkdir(parents=True, exist_ok=True)
+  output_path.write_text(output, encoding="utf-8")
 
 
 if __name__ == "__main__":
-    main()
+  main()

package/bundled-skills/wechat-assistant/skill/scripts/wechat_html_api_publish.py

@@ -0,0 +1,288 @@
+#!/usr/bin/env python3
+"""Create WeChat Official Account draft from markdown or rendered HTML.
+
+Security notes:
+- Credentials are NOT hardcoded in this script.
+- Provide appid/appsecret via args, env vars, or local config file.
+"""
+
+from __future__ import annotations
+
+import argparse
+import json
+import mimetypes
+import re
+import sys
+from pathlib import Path
+from typing import Dict, Optional
+from urllib import error, parse, request
+
+from render_wechat_api_content import render_api_content
+
+
+def escape_attr(text: str) -> str:
+  return text.replace("&", "&amp;").replace("<", "&lt;").replace(">", "&gt;").replace('"', "&quot;")
+
+
+def strip_tags(text: str) -> str:
+  text = re.sub(r"<[^>]+>", "", text)
+  text = text.replace("&nbsp;", " ")
+  return re.sub(r"\s+", " ", text).strip()
+
+
+def extract_title_from_html(full_html: str, fallback: str = "公众号文章") -> str:
+  match = re.search(r'<h1[^>]*class="[^"]*wx-title[^"]*"[^>]*>(.*?)</h1>', full_html, re.I | re.S)
+  if match:
+    title = strip_tags(match.group(1)).strip()
+    if title:
+      return title
+  match = re.search(r"<title>(.*?)</title>", full_html, re.I | re.S)
+  if match:
+    title = strip_tags(match.group(1)).strip()
+    if title:
+      return title
+  return fallback
+
+
+def extract_main_content_from_html(full_html: str) -> str:
+  match = re.search(r'<section[^>]*class="[^"]*wx-content[^"]*"[^>]*>([\s\S]*?)</section>', full_html, re.I)
+  if not match:
+    raise ValueError("无法从 HTML 中提取 .wx-content 内容，请检查输入文件。")
+  return match.group(1).strip()
+
+
+def iter_local_img_sources(content_html: str) -> list[str]:
+  return re.findall(r'<img[^>]+src="([^"]+)"', content_html, re.I)
+
+
+def normalize_local_asset_path(src: str, html_path: Path) -> Optional[Path]:
+  if re.match(r"^https?://", src, re.I) or src.startswith("data:"):
+    return None
+  p = Path(src)
+  if not p.is_absolute():
+    p = (html_path.parent / p).resolve()
+  return p
+
+
+def replace_inline_images(content_html: str, mapping: Dict[str, str]) -> str:
+  out = content_html
+  for old, new in mapping.items():
+    out = out.replace(f'src="{old}"', f'src="{escape_attr(new)}"')
+  return out
+
+
+def http_json(url: str, payload: Optional[dict] = None, method: str = "GET") -> dict:
+  data = None
+  headers = {"User-Agent": "OpenSchool-WeChat-Draft/1.0"}
+  if payload is not None:
+    data = json.dumps(payload, ensure_ascii=False).encode("utf-8")
+    headers["Content-Type"] = "application/json; charset=utf-8"
+
+  req = request.Request(url, data=data, headers=headers, method=method)
+  try:
+    with request.urlopen(req, timeout=60) as resp:
+      return json.loads(resp.read().decode("utf-8"))
+  except error.HTTPError as exc:
+    detail = exc.read().decode("utf-8", errors="replace")
+    raise RuntimeError(f"HTTP {exc.code}: {detail}") from exc
+
+
+def http_upload_file(url: str, file_path: Path, field_name: str = "media") -> dict:
+  boundary = "----OpenSchoolBoundary7MA4YWxkTrZu0gW"
+  mime_type = mimetypes.guess_type(file_path.name)[0] or "application/octet-stream"
+
+  parts: list[bytes] = []
+  parts.append(f"--{boundary}\r\n".encode())
+  parts.append(
+    (
+      f'Content-Disposition: form-data; name="{field_name}"; filename="{file_path.name}"\r\n'
+      f"Content-Type: {mime_type}\r\n\r\n"
+    ).encode()
+  )
+  parts.append(file_path.read_bytes())
+  parts.append(f"\r\n--{boundary}--\r\n".encode())
+  data = b"".join(parts)
+
+  req = request.Request(
+    url,
+    data=data,
+    headers={
+      "Content-Type": f"multipart/form-data; boundary={boundary}",
+      "User-Agent": "OpenSchool-WeChat-Draft/1.0",
+    },
+    method="POST",
+  )
+  try:
+    with request.urlopen(req, timeout=120) as resp:
+      return json.loads(resp.read().decode("utf-8"))
+  except error.HTTPError as exc:
+    detail = exc.read().decode("utf-8", errors="replace")
+    raise RuntimeError(f"HTTP {exc.code}: {detail}") from exc
+
+
+def get_access_token(appid: str, appsecret: str) -> str:
+  url = (
+    "https://api.weixin.qq.com/cgi-bin/token?grant_type=client_credential"
+    f"&appid={parse.quote(appid)}&secret={parse.quote(appsecret)}"
+  )
+  res = http_json(url)
+  token = res.get("access_token")
+  if not token:
+    raise RuntimeError(f"获取 access_token 失败: {json.dumps(res, ensure_ascii=False)}")
+  return token
+
+
+def upload_inline_image(access_token: str, file_path: Path) -> str:
+  url = f"https://api.weixin.qq.com/cgi-bin/media/uploadimg?access_token={parse.quote(access_token)}"
+  res = http_upload_file(url, file_path)
+  remote = res.get("url")
+  if not remote:
+    raise RuntimeError(f"上传正文图片失败: {json.dumps(res, ensure_ascii=False)}")
+  return remote
+
+
+def upload_cover_thumb(access_token: str, file_path: Path) -> str:
+  url = f"https://api.weixin.qq.com/cgi-bin/material/add_material?access_token={parse.quote(access_token)}&type=image"
+  res = http_upload_file(url, file_path)
+  media_id = res.get("media_id")
+  if not media_id:
+    raise RuntimeError(f"上传封面失败: {json.dumps(res, ensure_ascii=False)}")
+  return media_id
+
+
+def create_draft(access_token: str, article: dict) -> dict:
+  url = f"https://api.weixin.qq.com/cgi-bin/draft/add?access_token={parse.quote(access_token)}"
+  res = http_json(url, payload={"articles": [article]}, method="POST")
+  if res.get("errcode", 0) not in (0, None):
+    raise RuntimeError(f"创建草稿失败: {json.dumps(res, ensure_ascii=False)}")
+  return res
+
+
+def load_config(path: Optional[str]) -> dict:
+  if not path:
+    return {}
+  cfg_path = Path(path)
+  if not cfg_path.exists():
+    raise FileNotFoundError(f"配置文件不存在: {cfg_path}")
+  return json.loads(cfg_path.read_text(encoding="utf-8"))
+
+
+def main() -> int:
+  parser = argparse.ArgumentParser(description="Publish WeChat draft via Official Account API")
+  parser.add_argument("--markdown", default="", help="Markdown article path")
+  parser.add_argument("--html", default="", help="Rendered HTML path")
+  parser.add_argument("--config", default="", help="Path to wechat publish config json")
+  parser.add_argument("--appid", default="", help="WeChat appid (optional, overrides config)")
+  parser.add_argument("--appsecret", default="", help="WeChat appsecret (optional, overrides config)")
+  parser.add_argument("--title", default="", help="Override article title")
+  parser.add_argument("--author", default="", help="Override article author")
+  parser.add_argument("--digest", default="", help="Override article digest")
+  parser.add_argument("--cover", default="", help="Cover image path")
+  parser.add_argument("--content-source-url", default="", help="Source url in draft")
+  parser.add_argument("--need-open-comment", type=int, default=0)
+  parser.add_argument("--only-fans-can-comment", type=int, default=0)
+  parser.add_argument("--output-json", default="", help="Output result json path")
+  parser.add_argument("--dry-run", action="store_true", help="Build payload only")
+  args = parser.parse_args()
+
+  if not args.markdown and not args.html:
+    raise RuntimeError("必须提供 --markdown 或 --html 之一。")
+
+  config = load_config(args.config)
+  appid = args.appid or config.get("appid", "")
+  appsecret = args.appsecret or config.get("appsecret", "")
+  author = args.author or config.get("author", "")
+  content_source_url = args.content_source_url or config.get("content_source_url", "")
+
+  if args.markdown:
+    markdown = Path(args.markdown).read_text(encoding="utf-8")
+    title, content_html, _ = render_api_content(markdown, args.title)
+    html_path_for_assets = Path(args.markdown)
+  else:
+    html_text = Path(args.html).read_text(encoding="utf-8")
+    title = args.title or extract_title_from_html(html_text)
+    content_html = extract_main_content_from_html(html_text)
+    html_path_for_assets = Path(args.html)
+
+  digest = args.digest or strip_tags(content_html)[:120]
+  image_mapping: Dict[str, str] = {}
+
+  access_token = ""
+  if not args.dry_run and (iter_local_img_sources(content_html) or args.cover):
+    if not appid or not appsecret:
+      raise RuntimeError("需要提供 appid/appsecret 才能上传图片与创建草稿。")
+    access_token = get_access_token(appid, appsecret)
+
+  for src in iter_local_img_sources(content_html):
+    local_path = normalize_local_asset_path(src, html_path_for_assets)
+    if local_path is None:
+      continue
+    if not local_path.exists():
+      raise FileNotFoundError(f"正文图片不存在: {local_path}")
+    if args.dry_run:
+      image_mapping[src] = f"https://example.invalid/mock/{local_path.name}"
+    else:
+      image_mapping[src] = upload_inline_image(access_token, local_path)
+
+  if image_mapping:
+    content_html = replace_inline_images(content_html, image_mapping)
+
+  thumb_media_id = ""
+  if args.cover:
+    cover_path = Path(args.cover)
+    if not cover_path.exists():
+      raise FileNotFoundError(f"封面图不存在: {cover_path}")
+    if args.dry_run:
+      thumb_media_id = "MOCK_THUMB_MEDIA_ID"
+    else:
+      if not access_token:
+        if not appid or not appsecret:
+          raise RuntimeError("需要提供 appid/appsecret 才能上传封面。")
+        access_token = get_access_token(appid, appsecret)
+      thumb_media_id = upload_cover_thumb(access_token, cover_path)
+
+  article = {
+    "title": title,
+    "author": author,
+    "digest": digest,
+    "content": content_html,
+    "thumb_media_id": thumb_media_id,
+    "content_source_url": content_source_url,
+    "need_open_comment": args.need_open_comment,
+    "only_fans_can_comment": args.only_fans_can_comment,
+  }
+
+  if args.dry_run:
+    result = {
+      "mode": "dry-run",
+      "article": article,
+      "inline_image_replacements": image_mapping,
+    }
+  else:
+    if not appid or not appsecret:
+      raise RuntimeError("需要提供 appid/appsecret 才能创建草稿。")
+    if not access_token:
+      access_token = get_access_token(appid, appsecret)
+    result = create_draft(access_token, article)
+    result["article_preview"] = {
+      "title": title,
+      "author": author,
+      "digest": digest,
+      "thumb_media_id": thumb_media_id,
+      "inline_image_replacements": image_mapping,
+    }
+
+  rendered = json.dumps(result, ensure_ascii=False, indent=2)
+  if args.output_json:
+    output_path = Path(args.output_json)
+    output_path.parent.mkdir(parents=True, exist_ok=True)
+    output_path.write_text(rendered, encoding="utf-8")
+  try:
+    print(rendered)
+  except UnicodeEncodeError:
+    sys.stdout.buffer.write((rendered + "\n").encode("utf-8"))
+  return 0
+
+
+if __name__ == "__main__":
+  raise SystemExit(main())

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@openschool_01/skills",
-  "version": "0.1.3",
+  "version": "0.1.4",
   "description": "OpenSchool curated skills installer CLI.",
   "license": "MIT",
   "bin": {

package/registry/skills-registry.json

@@ -3,7 +3,7 @@
     "slug": "clawdhub",
     "name": "clawdhub",
     "tagline": "技能市场入口，帮你搜索、安装、更新和发布技能。",
-    "category": "starter",
+    "category": "office",
     "badge": "OpenSchool 推荐",
     "status": "已上线",
     "difficulty": "入门",
@@ -282,7 +282,7 @@
     "slug": "summarize",
     "name": "summarize",
     "tagline": "快速总结网页、视频、PDF 等内容，方便继续加工。",
-    "category": "starter",
+    "category": "office",
     "badge": "OpenSchool 推荐",
     "status": "已上线",
     "difficulty": "入门",