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 instructions/voidvec/drogon-claude-plugin/claude-mdgit clone --depth 1 https://github.com/voidvec/drogon-claude-pluginWhat 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 | $0.01799 | $0.01799 |
| Opus 5 | $0.00899 | $0.00899 |
| Sonnet 5 | $0.00360 | $0.00360 |
| Haiku 4.5 | $0.00180 | $0.00180 |
Grade A, and why
drogon-claude-plugin CLAUDE.md 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 yesterday.
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 — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Drogon 后端开发规则
本文件由
drogonClaude Code 插件提供。规则针对 drogon v1.9.13。本文只列顶层纪律（跨所有任务的高频陷阱）+ Skill 路由表。具体模板、API 速查、配置格式等详细知识已下沉到各 Skill 的
references/code-guide.md，按需加载——遇到对应任务时主动调用对应 Skill，不要凭记忆写代码。
A. 异步回调模型（所有 handler 的最高纪律）
Drogon 的每个 handler 接收一个 std::function<void(const HttpResponsePtr &)> 回调。响应的发出只与回调是否被调用有关，与 handler 是否 return 无关。
- 恰好一次：每个 handler 必须在所有代码路径（含错误/异常/提前返回）上恰好调用一次
callback。 - 绝不重复：禁止第二次调用
callback（未定义行为，可能重复响应或连接污染）。 - 禁止 handler 内同步阻塞：不得在 handler 内做阻塞式同步 I/O（同步读 DB、
sleep、长计算、阻塞文件读写）后再callback。改用 drogon 异步 API，或丢到线程池，完成处就近callback。 - 按值捕获回调：异步/延迟场景下
callback必须按值捕获进闭包，确保生命周期超出 handler 栈帧；悬空回调会崩溃。 - 异常不得逃逸 handler：用 try/catch 包裹业务逻辑，每个 catch 路径都调
callback(错误响应)。框架默认异常处理器只是兜底，不能替代你对 callback 的保证。 - 优先协程：
__cpp_impl_coroutine已定义且 CMakeUSE_COROUTINE=ON时，优先用协程 handler（Task<HttpResponsePtr>，框架负责异常与响应交付）。AsyncTask抛未处理异常会std::terminate，用时必须 try/catch。协程细节用drogon-gen-coroutine-handlerskill。
B. 事件循环模型（Trantor IO）
Drogon 跑在 trantor 的事件循环上；配置项 number_of_threads 决定循环数（值为 0 表示按 CPU 硬件并发数自动设置），每条连接钉在某个循环上。
- 事件循环线程不可阻塞：禁止在事件循环线程做任何阻塞工作（同步 DB 调用、
sleep、长计算、阻塞文件 I/O）——会拖死同一循环上所有其他连接。 - 重/阻塞活儿走线程池：把重活儿交给独立线程池；完成后用
runInLoop派回连接所属的事件循环，再在那里callback。 - 跨循环共享状态加锁：跨循环共享的状态必须加锁；更优是保持状态"循环局部"，或用
runInLoop派发到归属循环访问。
Skill 路由表
遇到下列任务时，主动调用对应 Skill——详细模板、API 签名、禁止模式清单都在 skill 的 references/code-guide.md 里，按需加载以节省上下文。
| 任务 | Skill | 说明 |
|---|---|---|
| 创建控制器（.h+.cc） | drogon-create-controller |
simple/http/websocket；含路径前缀差异、:param、WebSocket 宏、自动注册、生命周期纪律 |
| 现代 lambda 内联路由 | drogon-gen-lambda-handler |
app().registerHandler + {N} 参数绑定；与经典宏二选一不混用 |
| ORM CRUD 代码 | drogon-gen-orm-crud |
回调式 + 协程式；含禁 execSqlSync、双回调、事务纪律 |
| 数据库配置 | drogon-gen-db-config |
PG/MySQL/SQLite；含键名黑名单、loadConfigFile 异常、运行期 DrogonDbException、SQL 注入防护 |
| Redis 配置 + 用法 | drogon-gen-redis-config |
配置格式 + 单例/异步/异常/订阅防泄漏运行期纪律 |
| 通用配置文件 | drogon-setup-config |
listeners/app/SSL/会话；含路径解析、加载异常、键名黑名单、多环境 |
| CMake 构建 | drogon-gen-cmake |
find_package/ORM/协程；含 Conan、插件过滤器编译、禁硬编码路径 |
| 协程 handler/中间件 | drogon-gen-coroutine-handler |
Task/AsyncTask 语义、参数按值生命周期、forwardCoro |
| HTTP 出站请求 | drogon-gen-http-client |
异步/协程/反向代理 forward；含同步 sendRequest 死锁保护、timeout |
| CSP 视图模板 | drogon-gen-csp-view |
语法速查、布局、HttpViewData；含 drogon_ctl create view 管线、禁拼 HTML |
| Filter（请求拦截） | drogon-gen-filter |
认证/限流/输入校验；含 doFilter(fcb,fccb)、AutoCreation=false |
| Middleware（全局处理链） | drogon-gen-middleware |
日志/CORS/性能计时；含 invoke(nextCb,mcb) |
| Plugin（系统级扩展） | drogon-gen-plugin |
连接池/第三方 SDK 初始化；含三者职责边界对比 |
| Advice（AOP 切面） | drogon-gen-advice |
11 个内建切面；含拦截型 vs 观察型、注册时机 |
| 文件上传 | drogon-gen-file-upload |
MultiPart 解析+校验+落盘；含 setClientMaxBodySize、路径穿越 |
| Session 登录鉴权 | drogon-gen-session-auth |
登录/登出/鉴权；含 changeSessionIdToClient 防 fixation |
| 测试 | drogon-gen-test |
DROGON_TEST；含断言宏、异步测试、CMake 扫描、addDbClient |
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.
- yesterday First seen · 61 lines · 1,799 tokens per session scan A f4eae212034d
drogon-claude-plugin CLAUDE.md is an instructions file published in the GitHub repository voidvec/drogon-claude-plugin (1 stars, last pushed 5d ago), licensed MIT. It adds 1,799 tokens to every session, about $0.0090 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 2026-08-31.
Other instructions, from other repositories
scaffold AGENTS.md
Instructions for zigrivers/scaffold, covering agent instructions, quick reference, scaffold releases and landing the plane (session completion).
scaffold CLAUDE.md
Instructions for zigrivers/scaffold, covering claude.md, core principles, project overview, structure and source of truth.
sdd CLAUDE.md
Claude Code instructions for LiorCohen/sdd, covering claude.md, task management rules (mandatory), output rules, git rules and build rules.
claude-code-templates CLAUDE.md
Instructions for Justdvp/claude-code-templates, covering claude.md, project overview, development commands, package management and application commands.
codexspec CLAUDE.md
Instructions for Zts0hg/codexspec, covering claude.md - codexspec development guide, project overview, purpose, technology stack and architecture.
trellis AGENTS.md
Instructions for craigcossairt/trellis, covering agents.md , project, what i need from agents, current state - source of truth pointers and tech stack.