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 agentmods add skills/nongjun/feishu-cursor-claw/create-skillnpx skills add nongjun/feishu-cursor-claw --skill create-skillgit clone --depth 1 https://github.com/nongjun/feishu-cursor-clawWrote 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/nongjun/feishu-cursor-claw/create-skill)<a href="https://agentmods.dev/skills/nongjun/feishu-cursor-claw/create-skill"><img src="https://agentmods.dev/badge/skills/nongjun/feishu-cursor-claw/create-skill.svg" alt="Measured on agentmods" 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 | $0.00037 | $0.01484 |
| Opus 5 | $0.00018 | $0.00742 |
| Sonnet 5 | $0.00007 | $0.00297 |
| Haiku 4.5 | $0.00004 | $0.00148 |
Grade A, and why
create-skill 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 5d 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 — 195 lines — stays where its author put it; the contents beside it link to each section on GitHub.
创建 Cursor Agent Skill
开始之前：收集需求
- 目的和范围：这个 skill 帮助完成什么任务？
- 存放位置：个人 skill（
~/.cursor/skills/）还是项目 skill（.cursor/skills/）？ - 触发场景：什么时候 agent 应该使用这个 skill？
- 领域知识：agent 需要什么它本身不知道的专业信息？
- 输出格式：有没有特定的模板或样式要求？
如果有 AskQuestion 工具可用，用结构化问题收集；否则直接对话式询问。
目录结构
skill-name/
├── SKILL.md # 必须 — 主指令文件
├── reference.md # 可选 — 详细文档
├── examples.md # 可选 — 使用示例
└── scripts/ # 可选 — 工具脚本
└── helper.sh
存放路径
| 类型 | 路径 | 作用域 |
|---|---|---|
| 个人 | ~/.cursor/skills/skill-name/ |
所有项目可用 |
| 项目 | .cursor/skills/skill-name/ |
跟随仓库共享 |
禁止在 ~/.cursor/skills-cursor/ 下创建，该目录为 Cursor 内置技能保留。
SKILL.md 格式
---
name: your-skill-name
description: 简要描述做什么以及何时使用
---
# Skill 标题
## 使用方法
清晰的分步指导。
## 示例
具体的使用范例。
元数据字段
| 字段 | 要求 | 用途 |
|---|---|---|
name |
最长 64 字符，仅小写字母/数字/连字符 | 唯一标识 |
description |
最长 1024 字符，非空 | agent 据此决定是否应用该 skill |
写好 description
description 是 skill 被发现和应用的关键。agent 根据它决定何时使用。
原则
-
第三人称（description 会被注入系统提示词）：
- 好：
"处理 Excel 文件并生成报表" - 差：
"我可以帮你处理 Excel 文件"
- 好：
-
具体 + 包含触发词：
- 好：
"从 PDF 提取文字和表格，填写表单，合并文档。当处理 PDF 文件时使用。" - 差：
"帮助处理文档"
- 好：
-
同时包含 WHAT 和 WHEN：
- WHAT：skill 做什么（具体能力）
- WHEN：什么时候应该用（触发场景）
核心写作原则
1. 简洁为王
上下文窗口与对话历史、其他 skill、请求共享。每个 token 都在竞争空间。
默认假设：agent 已经很聪明。只添加它不知道的上下文。
2. SKILL.md 控制在 500 行以内
详细参考放入单独文件，agent 需要时再读取。
## 更多资源
- API 详情见 [reference.md](reference.md)
- 使用示例见 [examples.md](examples.md)
引用保持一层深度 — 从 SKILL.md 直接链接到参考文件。
3. 匹配自由度
| 自由度 | 何时使用 | 示例 |
|---|---|---|
| 高（文字指导） | 多种有效方案，依赖上下文 | 代码审查指南 |
| 中（伪代码/模板） | 有首选模式但允许变通 | 报告生成 |
| 低（具体脚本） | 脆弱操作，一致性关键 | 数据库迁移 |
常用模式
模板模式 — 提供输出格式
## 报告结构
\`\`\`markdown
# [标题]
## 摘要
## 关键发现
## 建议
\`\`\`
示例模式 — 靠示例保证输出质量
## 提交消息格式
**示例 1：**
输入：添加了 JWT 用户认证
输出：`feat(auth): implement JWT-based authentication`
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.
- 5d ago First seen · 195 lines · 37 tokens per session scan A a95c07a63eb8
create-skill is a skill published in the GitHub repository nongjun/feishu-cursor-claw (14 stars, last pushed 1mo ago), licensed MIT. It adds 37 tokens to every session and 1,484 once invoked, about $0.0002 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
metabot-team
Use for the MetaBot Agent Teams CLI: create and inspect durable Teams, spawn teammates, exchange messages, and manage Tasks and Runs across Sessions.
firebase-messaging
Use when setting up Firebase Cloud Messaging, managing permissions and tokens, handling background/foreground notification taps, or dispatching messages server-side (HTTP v1).
firebase-cloud-functions
Use when calling callable functions (httpsCallable), passing data to server-side logic, handling function errors/timeouts, configuring regions, or testing with the Emulator Suite.
openlark-api
OpenLark 项目 API 接口实现规范（速查）。用于添加/重构飞书开放平台 API：确定落盘路径、实现 Body/Response + Builder(Request)、对齐 endpoints 常量/enum、补齐 mod.rs 导出，并明确"调用服务端 API"的方法签名/RequestOption 传递约定。触发关键词：API 接口、API 文件、飞书 API、添加 API、调用服务端 API.
openlark-api-validation
OpenLark API 覆盖率验证技能。用于验证各 crate 的 API 实现数量与覆盖率，基于 tools/validateapis.py 脚本和 apilistexport.csv 对比实际代码实现。触发关键词：API 验证、API 覆盖率、验证 API 数量、检查 API 实现、API 统计.
openlark-design-review
OpenLark Rust SDK 的代码设计/公共 API 规范审查技能（面向 crate/模块）。用于系统化检查入口设计、feature gating、Request/Service/Builder 一致性、端点体系、Config/错误处理、导出与文档同步、测试与告警控制，并输出按优先级排序的整改清单与可落地改造方案。触发关键词：设计审查、crate 设计、API 设计、public API、feature flag、端点、Builder/Service、架构一致性.