Aquarius-mu

每日 AI 行业新闻聚合工具。从 30+ 白名单源(RSS/Atom/JSON API/Web)采集 AI 新闻,经主 Agent 评分筛选后,通过飞书群机器人推送 Interactive Card 卡片;未配置飞书 Bot 时直接在对话窗口 输出 Markdown 摘要。内置自我进化引擎: 自动发现并注册新信源(信源自我拓展)+ 按历史表现 动态调整信源品质信用分(品质自我进化)。当用户提到 AI日报、AI每日推送、今日AI新闻、 AI Daily Pulse、ai-daily-pulse 时使用此 Skill。

小红书推文封面生成器（极简独立杂志 Zine 风格）。用纯代码绘制 3:4 竖版封面： 浅灰蓝旧纸背景、纸张纤维、扫描颗粒、撕纸边缘、Risograph 印刷颗粒、轻微套色偏移。 默认用于「模型大评测」系列（大标题 + 本期模型 + 六张竖排撕边纸条模型名）， 文字 100% 拼写准确、风格永远统一。当用户提到 模型大评测封面、小红书封面、 zine 风格封面、推文配图、系列封面 时使用。

Build websites and web UI with smooth, polished, high-end motion by default — the buttery feel of premium sites, not static slop. Use when building or restyling any website, landing page, hero, marketing page, portfolio, or web UI; when adding scroll…

游戏后端功能的对抗性业务验收。以「严厉策划 + QA + 行为不定且渴望正常反馈的玩家」三视角，运用反馈路径表、玩家旅程时间线、对抗场景卡、状态机审计、消息契约审计、错误码体验审计、状态同步审计等多种技术，逐分支验证服务端在每个时机是否向客户端发出完整、正确、及时的反馈（回包 / 状态同步 / 主动推送 / 错误码 / 限制拦截），专门揪出"玩家做了操作却看不到正确反馈、或看到错误反馈"的功能向…

Use at the start of any session to understand which skill to invoke. Maps the full workflow — feature development, bug fixing, architecture, and utilities — for a C++ game server project with SVN.

Disciplined implement→build→review→commit loop for C++ game server features. Use when executing a plan task, fixing a bug, or making any code change. One vertical slice at a time.

Use when you have a spec or requirements for a multi-step task, before touching code. Writes a step-by-step implementation plan with exact file paths, code, and build commands.

Build a throwaway prototype to answer a design question before committing to an implementation. Use when sanity-checking a data model, state machine, or design approach.

Find deepening opportunities in the codebase — refactors that reduce coupling, simplify interfaces, and improve AI-navigability. Use when improving architecture, finding refactoring opportunities, or making a codebase easier to work with.

Use before any feature work — explores intent, asks clarifying questions one at a time, produces a design doc. Invoke before writing any code or plan.

Ultra-compressed communication mode. Cuts token usage ~75% by dropping filler, articles, and pleasantries while keeping full technical accuracy. Use when user says "caveman mode", "talk like caveman", "less tokens", or invokes /caveman.

Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → review. Use when something is broken, throwing errors, or regressing.

Interview the user or a design relentlessly until reaching shared understanding. Walk down every branch of the decision tree, resolving dependencies one by one. Use when stress-testing a plan, clarifying design, or the user says "grill me".

Use at the start of any session to understand which skill to invoke. Maps the full workflow — feature development, bug fixing, architecture, and utilities — for a C++ game server project with SVN.

Compact the current conversation into a handoff document for another agent or session to pick up.

Disciplined implement→build→review→commit loop for C++ game server features. Use when executing a plan task, fixing a bug, or making any code change. One vertical slice at a time.

飞书交互卡片（schema 2.0）：构建和发送卡片消息，含 @提及、流式思考态、原地更新、按钮/选择器交互。通过 lark-cli --as bot 发送，无需手动管理 token。

飞书Bot文件操作安全规范：白名单写入 ~/feishu_bot/cache/，禁止操作敏感路径和脚本文件，所有写入必须 flock+jq 原子替换。

飞书 lark-cli 操作完整指南：消息/文档/表格/群组/用户/日历/任务，内嵌所有关键规则和典型流程。

飞书Bot缓存管理：读写 ~/feishu_bot/cache/ 下的 tokens/groups/users JSON，含TTL校验、flock并发写入、name_cache同步、旧cache.json迁移。

国内网络环境 GitHub 下载加速工具。通过 SSH 端口 443 克隆仓库，通过镜像代理下载 Release 文件，绕过 GFW 限制。适用场景：(1) 克隆 GitHub 仓库失败或极慢；(2) 下载 Release 文件超时；(3) 使用国内云服务器（腾讯云、阿里云等）需要访问 GitHub。