Smart Web Monitor (智能网页监控) — 智能网页监控（Smart Web Monitor）

Smart 网页 监控 (智能网页监控)

Watch any 网页 page for changes — with AI-powered understanding. Instead of rigid keyword matching, describe what you're looking for in natural language and let the 代理 decide.

Why "Smart"? 普通监控 Smart 网页 监控 关键词匹配：risk 管理器 自然语言："是否有风控岗位？排除实习" 静态正则：risk.{0,20}管理器 代理 理解上下文语义 匹配即通知 匹配后 代理 可以总结、分析、过滤 一次判断 代理 持续推理，理解页面结构变化 Quick 启动 创建 a 监控 # Simple: keyword match python3 scripts/监控.py 创建 \ --name "hk-risk-jobs" \ --url "https://careers.example.com/risk" \ --match-type keyword \ --match-value "risk 管理器,hong kong,合规"

# Smart: LLM match (代理 evaluates with natural language understanding) python3 scripts/监控.py 创建 \ --name "hk-fintech-news" \ --url "https://36kr.com" \ --match-type llm \ --match-value "是否有香港金融科技、虚拟银行相关新闻？排除广告。"

运行 (keyword/regex/css/jsonpath) python3 scripts/监控.py 运行 --event hk-risk-jobs # → auto-matches, auto-暂停s on match

Fetch for LLM (提取 text only) python3 scripts/监控.py fetch --event hk-fintech-news # → 输出s JSON with page text for LLM 代理 to evaluate

After Match — 暂停 & 恢复 python3 scripts/监控.py 恢复 --event xxx # 继续 python3 scripts/监控.py 暂停 --event xxx # 暂停 python3 scripts/监控.py disable --event xxx # 停用 python3 scripts/监控.py enable --event xxx # 启用

Match Types Type How It Works Smart? Example keyword Comma-separated keywords (case-insensitive) ❌ risk 管理器,hong kong regex Regular expression pattern ❌ risk.{0,20}管理器 css CSS selector — matches if elements exist ❌ .job-列出ing .title jsonpath JSONPath for API 响应s ❌ $.data.jobs[].title llm 🤖 代理 evaluates with LLM reasoning ✅ 是否有风控岗位？排除实习 🤖 LLM Match (Smart Matching — 核心功能)

This is what makes the 监控 "smart". Instead of rigid pattern matching, the 代理 reads the page and thinks.

How it works:

监控.py fetch — fetches page, 提取s 清理 text → 输出s JSON Cron 代理 — reads the text, evaluates using its own LLM reasoning 代理 decides: does this match the condition? If matched → notifies you with a summary → auto-暂停s (won't spam) If not matched → silently skips

This enables natural language conditions that would be impossible with regex:

"是否有新的职位空缺？排除已见过的" "页面上有没有提到 Series B 或以上的融资？" "是否出现了与 AI 安全相关的文章？排除广告和推广内容" 状态 Machine active ──(match found)──→ 暂停d ──(恢复)──→ active │ ↑ └──(disable)──→ disabled ──(enable)──────────┘

监控 Config 格式化 { "name": "hk-risk-jobs", "description": "Hong Kong risk management jobs", "urls": [{"url": "https://careers.example.com/risk", "label": "Example"}], "match": { "type": "llm", "value": "是否有香港风控或合规岗位？排除实习。" }, "interval": 7200, "状态": "active", "enabled": true }

Cron Integration For keyword/regex/css/jsonpath 监控s: OpenClaw cron 添加 \ --name "监控: hk-risk-jobs" \ --cron "0 /2 " \ --tz "Asia/Hong_Kong" \ --会话 isolated \ --message "运行 监控: cd /home/node/.OpenClaw/workspace/技能s/网页-监控 && python3 scripts/监控.py 运行 --event hk-risk-jobs" \ --announce --channel discord --to "user:YOUR_ID" --light-上下文

For LLM 监控s: OpenClaw cron 添加 \ --name "监控: hk-fintech-news" \ --cron "0 /3 *" \ --tz "Asia/Hong_Kong" \ --会话 isolated \ --message 'You are a 网页 监控 代理. 运行: cd /home/node/.OpenClaw/workspace/技能s/网页-监控 && python3 scripts/监控.py fetch --event hk-fintech-news. Read the 输出 JSON. For each URL, evaluate if the page content matches the condition: "是否有香港金融科技相关新闻？排除广告。" If matched: 输出 a brief summary of what matched. If not matched: say "no match". If matched, also 运行: python3 scripts/监控.py 暂停 --event hk-fintech-news' \ --announce --channel discord --to "user:YOUR_ID" --light-上下文

All Commands Command Usage 创建 创建 new 监控 运行 --event X 运行 with built-in matching fetch --event X Fetch page text (for LLM) 列出 Show all 监控s show --event X Show config 恢复/暂停/disable/enable 状态 management 删除 --event X 删除 监控 添加-url 添加 URL to 监控