Feishu Public Table Monitor

监控公开可访问的飞书 Wiki/文档中指定章节下的价格表或模型表，检测版本日期、模型新增/删除、倍率与价格变化，并输出适合 Telegram/Markdown 的变更通知。适用于用户要求监控公开飞书表格、价格表、模型列表、倍率表、产品清单变动并推送提醒的场景。

Security Scan

The skill's code and runtime instructions match its stated purpose (monitoring publicly-accessible Feishu wiki tables and producing Markdown diffs); it requests no extra credentials or installs and the behavior visible in the script is consistent with that purpose.

Purpose & Capability

Name/description (monitor public Feishu tables for price/model changes) align with the included script and SKILL.md. The script fetches a provided public Feishu URL, extracts embedded page data, parses a targeted section/table, creates snapshots, and diffs them — all consistent with the stated purpose. The script requires the 'requests' package which is appropriate for HTTP fetching.

Instruction Scope

SKILL.md instructs the agent to run the included script against public Feishu URLs, create an initial baseline, and optionally run on cron. The instructions do not ask the agent to read unrelated files, access arbitrary environment variables, or transmit data to third-party endpoints. They only reference the script, section-title, state-dir, and printing or producing Markdown diffs (expected for the task).

Install Mechanism

No install spec is present; this is an instruction-only skill with a helper script. That minimizes install risk. The script dynamically imports the widely used 'requests' library at runtime and raises a clear error if missing — reasonable for a simple HTTP fetcher.

Credentials

The skill declares no required environment variables, no credentials, and no config paths. The script only issues outbound HTTP GETs to the user-provided Feishu URL. No sensitive system credentials or unrelated environment access are requested or used.

Persistence & Privilege

The skill is not forced-always (always: false) and uses the normal autonomous-invocation defaults. SKILL.md recommends using cron to run the script, which is appropriate for a monitor. There is no indication the skill modifies other skills or system-wide settings.

Assessment

This skill appears coherent and limited to monitoring public Feishu pages. Before installing or automating: (1) Ensure the URLs you monitor are truly public — do not supply private/internal pages or credentials to this script. (2) Review or run the script locally first to confirm it correctly finds the intended table (use --print-snapshot). (3) Choose a safe state directory (state-dir) and ensure file permissions are appropriate. (4) Ensure the runtime environment has the 'requests' package or install it in a controlled environment. (5) If you plan to let the agent run this periodically, run it in an environment/network segment that limits unintended outbound access to arbitrary URLs. If you want extra assurance, inspect the rest of the script (the diff/notification output and any file-write logic) to confirm it only writes local snapshot files and does not send data to external endpoints beyond the Feishu URL you provide.

用于监控公开可访问的飞书 Wiki/文档页面中的目标表格喵。

适用场景

当用户提出这些需求时使用本 skill 喵

• 监控公开飞书页面中的价格表变化
• 监控模型列表、倍率表、资费表、商品表变化
• 需要把变动内容整理成 Markdown 通知
• 需要为该监控生成可定时执行的脚本或 cron job

限制

• 仅适用于无需登录即可访问的飞书页面喵
• 当前脚本按“章节标题下的首个表格”定位目标表喵
• 如果页面结构大改 可能需要调整脚本喵

快速用法

脚本路径喵

• scripts/monitor_feishu_price_table.py

先抓取一次基线喵

python3 scripts/monitor_feishu_price_table.py \
  'https://example.feishu.cn/wiki/XXXX' \
  --section-title '三、模型列表与倍率价格表（所有模型可用）'

如果只想看当前解析结果喵

python3 scripts/monitor_feishu_price_table.py \
  'https://example.feishu.cn/wiki/XXXX' \
  --section-title '三、模型列表与倍率价格表（所有模型可用）' \
  --print-snapshot

兼容说明喵

• 支持旧版 clientVars: Object({...}) HTML 内嵌数据喵
• 支持 Object(JSON.parse("..."))、Object(decodeURIComponent("...")) 等包装格式喵
• 如果飞书返回 Object() 空壳，说明页面已切换到新版异步加载；脚本会给出清楚错误，不再抛误导性的 JSONDecodeError: Expecting value 喵
• 对 Object() 空壳页面，后续应改用飞书实际接口或浏览器自动化抓取加载后的表格喵

常用参数

• --section-title：要监控的章节标题喵
• --title：通知标题 文案默认可自定义喵
• --state-dir：状态目录 用于存放基线快照喵
• --print-snapshot：打印当前解析出来的表格快照喵

推荐工作流

1. 先用 --print-snapshot 确认表格抓对了喵
2. 再正常跑一次初始化基线喵
3. 然后把脚本挂到 cron 定时执行喵
4. 若输出是 NO_REPLY 就表示没变化喵
5. 若输出是 Markdown 文本 就直接发送给用户喵

与 OpenClaw cron 搭配

如果用户要“有变化就推送给我”喵

• 用本脚本生成差异输出喵
• 在 cron 的 agentTurn 里执行脚本喵
• 规则写成：
  • 输出 NO_REPLY 或 INIT_ONLY 时只回复 NO_REPLY
  • 其他输出原样发送喵

自定义建议

如果用户希望排版更像公告喵

• 用 --title 自定义标题
• 保持 Markdown 样式输出
• 需要更细分类时 可扩展为 涨价 降价 新增 下架 四段喵