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 nongjun/feishu-cursor-claw --skill doc-gardenergit 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/doc-gardener)<a href="https://agentmods.dev/skills/nongjun/feishu-cursor-claw/doc-gardener"><img src="https://agentmods.dev/badge/skills/nongjun/feishu-cursor-claw/doc-gardener/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/nongjun/feishu-cursor-claw/doc-gardener"><img src="https://agentmods.dev/badge/skills/nongjun/feishu-cursor-claw/doc-gardener.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 | $0.00096 | $0.01549 |
| Opus 5 | $0.00048 | $0.00775 |
| Sonnet 5 | $0.00019 | $0.00310 |
| Haiku 4.5 | $0.00010 | $0.00155 |
Grade A, and why
doc-gardener 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 9d 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 — 157 lines — stays where its author put it; the contents beside it link to each section on GitHub.
文档园丁
知识库建立后的唯一维护入口。包含三大能力：巡检、整理、审计。
子 agent 使用规范
所有涉及多文件、多模块的操作，必须使用 Task 工具启动子 agent 并行处理。
- 不指定 model 参数（继承父级模型）
- 每批最多 4 个并行子 agent
- 每个子 agent 负责一个独立范围（如一个模块、一类文档）
能力A：巡检
触发："文档巡检" / "跑一下文档园丁" / "检查文档" / "扫描文档"
第一步：运行扫描脚本
bash ~/.cursor/skills/doc-gardener/scripts/scan.sh <项目根目录>
脚本执行 13项检查：
| 范围 | 检查项 |
|---|---|
| 文档结构 | 关键文件存在、AGENTS.md行数、设计文档验证状态、索引同步、过期（>30天）、空目录 |
| 代码一致性 | 容器一致（docker ps vs 架构.md）、端口一致、凭据一致（.env vs 凭据文档）、模块目录一致、部署模式一致 |
| Git分析 | 代码变更但文档未更新的模块、文档统计 |
第二步：子 agent 并行修复
根据扫描结果，启动并行子 agent 分头处理：
| 问题类型 | 子 agent 任务 |
|---|---|
| 容器/端口/模块不一致 | 更新 架构.md 和 AGENTS.md 对应部分 |
| 凭据不一致 | 读取 .env 最新值，更新凭据文档 |
| 部署模式不一致 | 读取 docker-compose，更新 架构.md 标记 |
| 代码变更未同步 | 逐模块分析 git diff，判断是否需要更新文档 |
Agent 额外主动检查：
- 质量评分是否覆盖所有模块
文档/内的 TODO/FIXME 标记执行计划/进行中/是否有已完成但未归档的计划AGENTS.md知识库索引是否与文档/子目录同步
第三步：生成巡检报告
存入 文档/变更日志/文档巡检报告-YYYYMMDD.md。
能力B：整理散乱文档
触发："整理文档" / "整理XX模块的文档"
当项目下有散落在各处的旧文档时使用。
流程
第一步：扫描盘点（子 agent）
启动 1 个子 agent 递归扫描目标范围内所有 .md 文件，生成审计清单：
| 序号 | 当前路径 | 文件名 | 大小 | 最后修改 | 初步分类 | 处置建议 |
处置建议类型：迁移 / 合并 / 归档 / 废弃 / 更新后迁移
第二步：并行分类迁移（多个子 agent）
按审计清单结果，启动并行子 agent：
| 子 agent | 负责范围 |
|---|---|
| agent-A | 迁移设计文档类（补验证状态头部） |
| agent-B | 迁移参考资料类（API文档、数据库Schema等） |
| agent-C | 迁移凭据/配置类 + 产品规格类 |
| agent-D | 处理需合并/归档/废弃的文档 |
每个子 agent 的执行动作：
- 迁移：复制到
文档/对应子目录，补验证状态头部 - 合并：整合多份文档为一份，保留各自独特信息
- 归档：移至
文档/执行计划/已完成/ - 废弃：删除文件，在变更日志中记录
第三步：清理与更新索引
- 清理原目录下已迁移的文档
- 清除空目录
- 更新
AGENTS.md知识库索引 - 更新
架构.md（如发现架构信息需修正）
第四步：输出审计报告
# 文档审计报告 — [范围] — [日期]
## 统计
## 发现的问题
## 新增的隐性知识
能力C：审计模块旧文档
触发："审计XX模块的文档" / "梳理XX模块"
对特定模块下的旧文档进行深度审计，验证其与代码的一致性。
流程
第一步：定位模块文档
读取目标模块目录，找出所有 .md 文件和文档类子目录（规划文档/、文档/ 等）。
第二步：逐份验证（子 agent 并行，每个 agent 处理 3-5 份文档）
对每份文档：
- 读取全文
- 对照当前代码验证：接口/逻辑/配置是否与代码一致
- 判定验证状态（✅/⚠️/❌/🚧）
- 识别隐性知识（决策原因、踩坑记录等，这些需保留）
What ships with it
1 file 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.
- 9d ago First seen · 157 lines · 96 tokens per session scan A 1f434facd58e
doc-gardener is a skill published in the GitHub repository nongjun/feishu-cursor-claw (14 stars, last pushed 1mo ago), licensed MIT. It adds 96 tokens to every session and 1,549 once invoked, about $0.0005 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.
enhance-web-web3d
Add purposeful 3D/WebGL and scroll choreography to an existing site with Three.js/R3F, GSAP, or Motion. Use when "add 3D", "WebGL hero", "React Three Fiber", or "scroll-driven 3D". General UI polish → enhance-web-ui. Motion without 3D → enhance-motion.
mobile-rn-screen
Polish an existing React Native screen to feel intentional, native, and human-crafted. Use for "this screen looks off", "feels clunky on iOS", "Android version looks wrong", "jank when scrolling", "button is unreachable", or any RN-specific UX polish pass.
audit-payment-system
Read-only audit for payment/money-movement systems, scope-gated so a Stripe-Checkout site and an in-house ledger each see only relevant findings. Use when "audit payment system", "double charge / idempotency", "ledger / reconciliation", "webhook / 3DS / PCI". Mobile IAP → audit-monetization-iap.