Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add AmbroseX/larksnap --skill larksnap-fetchgit clone --depth 1 https://github.com/AmbroseX/larksnapWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/ambrosex/larksnap/larksnap-fetch)<a href="https://agentmods.dev/skills/ambrosex/larksnap/larksnap-fetch"><img src="https://agentmods.dev/badge/skills/ambrosex/larksnap/larksnap-fetch/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/ambrosex/larksnap/larksnap-fetch"><img src="https://agentmods.dev/badge/skills/ambrosex/larksnap/larksnap-fetch.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00111 | $0.01898 |
| Opus 5 | $0.00056 | $0.00949 |
| Sonnet 5 | $0.00022 | $0.00380 |
| Haiku 4.5 | $0.00011 | $0.00190 |
Grade A, and why
larksnap-fetch scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.
larksnap-fetch
在任意项目的 Claude Code 里「扔一个飞书链接 → 落到本地目录」。本技能不直接请求飞书,而是把任务 交给已登录的 larksnap 浏览器扩展(它持有 cookie 与导出引擎)。
CC ──fetch.mjs(一次性)──HTTP /command──▶ daemon ──WebSocket──▶ Chrome 扩展
▲ 127.0.0.1:19925 └ 后台开标签页跑导出
└ 流式回传进度/产物 ──▶ 解包到 <输出目录>/<文档名>/
本技能自包含:daemon 与协议代码已随技能打包在 scripts/bridge/ 里,fetch.mjs 拉起的是
技能内部那份 daemon,不依赖 larksnap 仓库的目录结构,因此可从任何项目调用。唯一的外部依赖是
那个装在 Chrome 里、已登录飞书的扩展(扩展无法塞进技能,只能在浏览器里加载一次,见 references/setup.md)。
每篇文档落到自己的子文件夹(以文档标题命名),不和其它文件平铺混在一起:
<输出目录>/
└── 无监督数据修复/ ← 文档标题命名的文件夹
├── 无监督数据修复.md
└── images/ ← 图片用相对路径内联引用
└── xxx.png
- 扩展是 WS 客户端,主动连出到本地 daemon —— 无 native messaging、无系统级安装。
- daemon 由本技能按需自动拉起、持久存活(空闲 30 分钟自退),只绑
127.0.0.1。 - fetch.mjs 一次性:跑完即退,不常驻。
用法
node ~/.claude/skills/larksnap-fetch/scripts/fetch.mjs <飞书链接> <输出目录> [--format md|pdf|html] [--profile <code>]
<输出目录>是父目录;产物会落到<输出目录>/<文档名>/子文件夹里(每篇文档独立成夹,互不混淆)。--format缺省md。md/html会把图片下载并以相对路径内联到该子文件夹的images/;pdf视租户能力而定。- 进度打到 stderr,结果路径打到 stdout(
✓ 已导出到后是该文档的子文件夹绝对路径)。 --profile <code>:当有多个浏览器 profile 同时连到 daemon 时,指定用哪一个(code 见扩展弹窗的 Profile,可点 Copy 复制)。只有一个时无需指定。
退出码:0 成功 ｜ 1 失败 ｜ 2 用法错 ｜ 3 需登录 ｜ 4 需授权域名 ｜ 5 桥接未就绪。
网页搜索
需要搜索资料、找参考链接或从搜索结果继续抓正文时，读取 references/search.md。
搜索与 fetch 共用本技能的 daemon 和 bridge，脚本入口是 scripts/search.mjs。
编辑飞书文档(写入 Markdown 内容)
需要新建、追加、插入、替换、删除或检索飞书文档块时，读取 references/edit.md。
编辑与 fetch 共用本技能的 daemon 和 bridge，脚本入口是 scripts/edit.mjs。
普通网页与小红书
链接不是飞书文档时,fetch.mjs 自动走普通网页管线(整页转 Markdown),命令不变,只支持 md,
图片保留外链。小红书有专属适配器,要求登录 + 带 xsec_token 的完整分享链接,图片会本地化。
抓普通网页失败、或要抓小红书时,读取 references/webpage.md。
arXiv 论文
arXiv 链接或裸 ID 不走 daemon/扩展,用独立脚本一次下载 PDF + HTML + Markdown:
node ~/.claude/skills/larksnap-fetch/scripts/arxiv.mjs <arXiv链接或ID> <输出目录> [--pdf-only|--html-only]
What ships with it
17 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/arxiv.md 1.5 KB
- references/edit.md 5.1 KB
- references/errors.md 3.2 KB
- references/search.md 1.7 KB
- references/setup.md 3.7 KB
- references/webpage.md 1.6 KB
- scripts/arxiv.mjs 9.6 KB runs code
- scripts/bridge/client.mjs 13 KB runs code
- scripts/bridge/daemon.mjs 17 KB runs code
- scripts/bridge/protocol.mjs 8.8 KB runs code
- scripts/bridge/video.mjs 22 KB runs code
- scripts/edit.mjs 23 KB runs code
- scripts/fetch.mjs 9.6 KB runs code
- scripts/search.mjs 8.7 KB runs code
- scripts/unzip.mjs 3.4 KB runs code
- scripts/vendor/entry.mjs 467 B runs code
- scripts/vendor/turndown.bundle.mjs 243 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 111 lines · 111 tokens per session scan A 2f8a48613e16
larksnap-fetch is a skill published in the GitHub repository AmbroseX/larksnap (272 stars, last pushed 3d ago), licensed Apache-2.0. It adds 111 tokens to every session and 1,898 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
documents
Read and write Lark documents - get content as markdown or blocks, create new documents, append content (text, headings, lists, code), list folders. Use when user asks about a Lark doc, wants to read/create/edit a document, or mentions a document URL/ID.
sheets
Read and query Lark Sheets (spreadsheets) - list sheets in a spreadsheet, read cell data. Use when user asks about a spreadsheet, wants to read data from a Lark sheet, or mentions a spreadsheet URL/ID.
minutes
Access Lark Minutes recordings - get metadata, export transcripts, download audio/video. Use when user asks about meeting recordings, transcripts, or minutes.
recipe-convert-markdown
Convert markdown to other formats using Universal IR.
recipe-collect-data
Collect data from Lark spreadsheet and convert to Notion database.
recipe-export-archive
Export all Notion docs for offline archive.