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 DropFan/claude-code-plugins --skill lark-sharedgit clone --depth 1 https://github.com/DropFan/claude-code-pluginsWrote 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/dropfan/claude-code-plugins/lark-shared)<a href="https://agentmods.dev/skills/dropfan/claude-code-plugins/lark-shared"><img src="https://agentmods.dev/badge/skills/dropfan/claude-code-plugins/lark-shared/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/dropfan/claude-code-plugins/lark-shared"><img src="https://agentmods.dev/badge/skills/dropfan/claude-code-plugins/lark-shared.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00049 | $0.01393 |
| Opus 5.5 | $0.00020 | $0.00557 |
| Sonnet 5.5 | $0.00010 | $0.00279 |
| Haiku 4.5 | $0.00005 | $0.00139 |
Grade A, and why
lark-shared 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 25d 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.
This is a copy
88% identical to lark-shared — 220 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.
⚙️ Cowork / Claude Desktop 执行环境说明（自动注入）
本技能依赖本地
lark-cli（@larksuite/cli，可用command -v lark-cli定位）及其~/.lark-cli登录态（应用密钥存于 macOS keychain）。在 Cowork 中运行任何
lark-cli命令时，必须在本地 macOS 上执行（使用 Desktop Commander 的start_process/interact_with_process，或其它本地 shell 工具），不要用隔离的 Linux 沙箱mcp__workspace__bash——沙箱里没有 lark-cli、也读不到 keychain。 执行前确保 npm 全局 bin 目录（npm prefix -g输出目录下的bin）在 PATH 中。（在 Claude Code 中可忽略本说明，lark-cli 在本机 shell 直接可用。）
lark-cli 共享规则
所有 lark-* skill 共享的底座：身份、认证、输出契约与高风险操作。
通用准则
-
调用前先确认用法：执行前读对应 reference 或跑
--help，别猜 flag 盲调。 -
身份决定你代表谁操作：
--as user代表用户本人（能看到、也能操作其日历、云空间/云盘/云存储等个人资源），--as bot代表应用自己，应用级操作，只能访问bot自己的资源，bot 查用户资源会返回空成功而非报错。动手前先搞清楚身份identity。身份模型和权限管理 →lark-shared-identity-and-permissions.md。 -
授权 / 配置类 URL 必须配二维码：当命令输出
verification_url、verification_uri_complete、console_url等 URL 字段时，必须用lark-cli auth qrcode生成并在回复中展示，URL 在前二维码在后；优先生成 PNG（--output），仅当用户明确要求时才使用 ASCII（--ascii）。URL 原样转发——不编解码、不加标点、不重拼 query，二维码和链接请一起展示给用户。 -
--format json（默认）下，判断成功用ok == true（或进程退出码 0），不要用code == 0：成功信封没有顶层code/msg字段，code只出现在错误信封的error内。按 OpenAPI 老格式{"code": 0, "msg": "ok"}判断会把所有成功调用误判为失败——封装写入类命令时尤其危险。JSON 输出契约 →lark-shared-output-contract.md。
安全规则
-
禁止输出密钥（appSecret、accessToken等）到终端明文。
-
写入/删除操作前必须确认用户意图。
-
目标命令支持
--dry-run时，用--dry-run预览危险请求。 -
退出码 10 是高风险确认门禁（
risk: "high-risk-write"），不是错误：停下 → 向用户确认（展示action、risk和关键参数）→ 取得用户显式同意后，将hint指出的确认 flag 追加到你原始 argv 的末尾后重试；绝不静默加确认 flag 绕过 →lark-shared-high-risk-approval.md。 -
文件路径只接受相对路径：
--file、--output、--output-dir、@file等路径参数只接受 cwd 下的相对路径，传绝对路径会报unsafe file path。数据输入（@file、大 JSON）优先用 stdin 传入，避免路径和转义问题。
Reference 强触发索引
命中任一触发条件时，MUST 在执行下一步前读取对应 reference。命中多条时按表中顺序读取，同一reference只读取一次。
What ships with it
6 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.
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.
- 25d ago Changed · -163 lines 34b3a52f6293
- 1mo ago First seen · 223 lines · 49 tokens per session scan A f176e7ade7bc
lark-shared is a skill published in the GitHub repository DropFan/claude-code-plugins (8 stars, last pushed 26d ago), licensed MIT. It adds 49 tokens to every session and 1,393 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. It is 88% identical to lark-shared, differing in 220 lines, and is treated as a copy.
Other skills, from other repositories
fly
Fly one thought to its destination across concurrent agents, on whatever the project and harness already have. Lands only when the user has accounted for every detail, never on an agent's say-so.
deepen
Hold an ask to the bar a top professional in its field would set — a deeper ask, never a bigger one. Use when an ask is clear but nothing yet names the sources, methods, target, or outside check the work will meet.
consult
Bring how named experts would think about the question. Use when the user wants expert thinking on a tradeoff, choice, or stuck problem — not a verdict.
setup
Use when installing or refreshing sound taste rules in a project — an explicit "sound setup", starting work in a repo with no .claude/sound/, or a taste-not-defined nudge. Also use to re-run after a stack change (new framework, new package in a monorepo).
auto-loop
TDD-based autonomous development loop with checkpoint recovery and observability changelog.
hook-template
Generate hook script from template. Use when adding a new hook, wiring a PreToolUse/PostToolUse/Stop/Notification hook, or scaffolding hook config for settings.json.