OpenClaw不依赖skill技能定时检测这些站点是否有新文章，并展示部分内容

教程／实战
1

信息订阅，而不是网页自动化

  • skill 体系不完整（有占位但没安装）
  • OpenClaw CLI 行为不一致（版本问题）
  • 用外部简单工具反而更稳定
  • :backhand_index_pointing_right: 主要是免token

最优建议不依赖 OpenClaw skill 完整 Python 脚本 + 去重 + 输出摘要 + 自动运行

环境

Ubuntu系统
 ├── Python3（系统自带）✔
 ├── OpenClaw（独立工具）✔
 └── pip受限制（PEP 668）✔

脚本

cat > /data/rss_check.py << 'EOF'
import feedparser
import json
import os
import re

FEEDS = [
    "https://zhujiwiki.com/feed",
    "https://affyun.com/feed",
    "https://iui.su/feed",
    "https://lala.im/feed",
    "https://www.appinn.com/feed",
    "https://www.dayanzai.me/feed",
    "https://www.iplaysoft.com/feed"
]

STATE_FILE = "/root/rss_seen.json"

# 读取历史
if os.path.exists(STATE_FILE):
    with open(STATE_FILE, "r") as f:
        seen = set(json.load(f))
    first_run = False
else:
    seen = set()
    first_run = True

new_seen = set(seen)
new_items = []

for url in FEEDS:
    d = feedparser.parse(url)

    for entry in d.entries[:5]:
        if entry.link not in seen:
            new_seen.add(entry.link)

            if not first_run:
                summary = re.sub('<.*?>', '', entry.get("summary", ""))[:100]

                new_items.append(
                    f"[{url}]\n标题: {entry.title}\n链接: {entry.link}\n摘要: {summary}\n"
                )

if new_items:
    print("\n".join(new_items))

with open(STATE_FILE, "w") as f:
    json.dump(list(new_seen), f)
EOF

执行脚本

python3 /data/rss_check.py