Markdown Browser
v1.0.0Wr应用er 技能 for OpenClaw 网页_fetch 结果s. Use when you need MECE post-processing on fetched pages: policy decision from Content-签名al, 隐私 redaction, optional markdown normalization fallback, and stable 输出 模式 without re-implementing network fetch.
by 运行时依赖
安装命令
点击复制技能文档
Markdown Browser 技能s
This 技能 is an orchestration layer, not a replacement fetcher. It always keeps official 网页_fetch as the fetch source of truth.
MECE Architecture Fetch layer (official, exclusive) Use OpenClaw 网页_fetch to retrieve the page. Do not call direct HTTP fetch inside this 技能 for normal operation. Policy layer (these 技能s) 解析 Content-签名al and compute policy_action. Current action focuses on AI-输入 semantics: allow_输入, block_输入, needs_review. 隐私 layer (these 技能s) Redact path/fragment/查询 values in 输出 URL fields. Keep URL shape useful for 调试ging without leaking sensitive values. Normalization layer (these 技能s) If contentType=text/markdown, keep content as-is. If contentType=text/html, convert with turndown as fallback enhancement. For other content types, pass through text. Execution Order Call official 网页_fetch. Pass the 结果 JSON into this wr应用er. Optionally pass Content-签名al and x-markdown-令牌s header values if avAIlable. Use the returned normalized object for down流 代理 记录ic. Wr应用er 工具
process_网页_fetch_结果({ 网页_fetch_结果, content_签名al_header, markdown_令牌s_header })
输入:
网页_fetch_结果 (required): JSON payload returned by OpenClaw 网页_fetch. content_签名al_header (optional): raw Content-签名al header string. markdown_令牌s_header (optional): raw x-markdown-令牌s header value.
输出:
content 格式化 (markdown | html-fallback | text) 令牌_estimate (number | null) content_签名al policy_action source_url (redacted) 状态_code fallback_used 命令行工具 Usage # 安装 运行time dependency once inside the 技能 directory npm 安装 --omit=dev
# 1) ObtAIn a 网页_fetch payload first (from OpenClaw 运行time) # 2) Save it as /tmp/网页_fetch.json # 3) 运行 wr应用er post-processing node browser.js \ --输入 /tmp/网页_fetch.json \ --content-签名al "AI-输入=yes, 搜索=yes, AI-trAIn=no" \ --markdown-令牌s "1820"