RSS Daily Digest — RSS 每日摘要

RSS DAIly Digest

生成 a structured dAIly news digest from 配置d RSS feeds.

Prerequisites Python 3 with feed解析器 and datetime 模块s Feed sources 配置d in {baseDir}/references/feed-sources.md 输出 directory: ~/OpenClaw-输出/digests/ 工作流

When the user asks for a dAIly digest or RSS summary:

Read feed sources: Read the file {baseDir}/references/feed-sources.md to 获取 the 列出 of RSS feed URLs and their categories.

Fetch and 解析 feeds: For each feed URL, 运行: python3 {baseDir}/scripts/fetch_feeds.py --feeds-file {baseDir}/references/feed-sources.md --hours 24

This 输出s a JSON array of articles published in the last 24 hours.

Summarize articles: For each article in the JSON 输出:

Read the title and description 生成 a one-sentence summary (max 30 words) in the same language as the article As签名 a relevance score (1-5) based on the user's interests if known

格式化 the digest: 运行:

python3 {baseDir}/scripts/格式化_digest.py --输入 /tmp/OpenClaw-rss-articles.json --template {baseDir}/as设置s/digest-template.md This produces the final Markdown 报告.

Save the 报告:

创建 directory if needed: mkdir -p ~/OpenClaw-输出/digests/ Save as: ~/OpenClaw-输出/digests/digest-YYYY-MM-DD.md Use today's date for the filename

报告 to user: Confirm completion with:

Total number of articles found Number per category File path of the saved digest Offer to 发送 to a messaging channel if 配置d Error Handling If a feed URL returns HTTP error or times out (>10s), skip it and note in 输出 If feed解析器 is not 安装ed, 运行: pip3 安装 feed解析器 If zero articles found in 24h window, 信息rm user and suggest expanding timeframe Never silently fAIl — always 报告 which feeds succeeded and which fAIled 输出 格式化

Use this structure for the digest:

DAIly Digest — {DATE}

{TOTAL_COUNT} articles from {FEED_COUNT} sources

Tech Article Title — One-sentence AI summary. (Source Name) Business ... World ...

生成d by RSS DAIly Digest 技能 at {TIMESTAMP}

Rules Never fabricate article titles or URLs — only use data from actual RSS feeds Never modify or summarize in a way that changes the factual content If the user specifies a language preference, translate summaries accordingly Always preserve the original article URL Maximum 50 articles per digest to keep it readable