longsizhuo

openInvest 多资产 AI 投资委员会 **日常使用**。读取持仓 / 实时行情 / 策略 / 历史决议 / 加减仓 / 跑 4 角色 LLM 委员会给投资 verdict。支持任意 yfinance symbol（A 股 / 港股 / 美股 / ETF / 加密 / 商品）和任意币种。**两条路径**：(1) Coordinator — Claude Code spawn 4 个 subagent，省 DeepSeek token；(2) Direct — 任何 agent（Cursor / Cline / Codex / 普通脚本）跑 `run.sh run_committee <SYM>` 一键拿 verdict。**触发场景**："show portfolio / 看看我的持仓"、"我现在涨了多少 / how is my P&L"、"该不该买/卖 X / should I buy X"、"分析一下 X / analyze X"、"跑委员会 / run committee on X"、"track AAPL / 跟踪苹果"、"加仓 / 减仓 / 记一笔交易"。**首次安装走另一个 skill `invest-setup`**（doctor 返回 needs_setup 时切过去）。后端 longsizhuo/openInvest。

备份 / 恢复 openInvest 的本地状态——memory/（持仓、策略、用户画像、委员会记录、dream 日志）+ db/（交易台账、job 运行历史、行情缓存）+ .env（SMTP/API 凭据）+ user_profile.json。这些数据全部 .gitignore，git 里没有任何历史版本，一旦被误覆盖（例如手滑直接跑了某个一次性迁移/初始化脚本）就是真实数据丢失，没有 git revert 可用。**主动触发场景**："备份一下 openInvest 的数据"、"backup invest data"、"我的持仓/策略好像被清空了"、"invest 数据丢了"、"恢复一下 invest 的备份"、"restore invest backup"、任何要在这台机器上重装/迁移 openInvest 部署之前、或者刚要跑一个陌生的迁移/初始化脚本之前（先备份再动手）。

First-time openInvest installation and onboarding. **ONLY use when** user explicitly says "set up invest" / "init invest" / "帮我初始化 invest", OR when `invest` skill's `doctor` returns `status: "needs_setup"`. **NOT for daily usage** — once onboarding is done, the `invest` skill takes over (portfolio viewing, committee analysis, buy/sell tracking). Wraps `run.sh init --from-stdin` with the canonical 5-question flow.

Maintain openInvest's docs (docs/wiki chapters + docs/wiki/adr) under Google's Open Knowledge Format (OKF). Two jobs. (1) Teach agents to maintain docs the OKF way — every doc carries a small YAML frontmatter block as the single source of truth (type, title, tags, intent, schema_source, documents); schema details link to the authoritative code instead of being copied into prose; no more hand-maintained thousand-line markdown. (2) Look docs up fast — grep the literal term FIRST; only when grep is ambiguous (hits scattered across files / synonym mismatch / zero hits) run find_docs.py to rank the owning doc by frontmatter intent, or resolve a doc's schema_source to the real code. Trigger phrases — "which doc covers X", "find the schema for PortfolioResponse", "where is GET /api/holdings documented", "docs for verdict.risk_profile", "add OKF frontmatter to this doc", "lint the wiki", "scaffold a new ADR/chapter". Run: scripts/run.sh find|schema|index|lint|new (or python3 scripts/find_docs.py --repo <path> ...).

帮助新手把 BossZhiPin_Job_Search 这个 BOSS 直聘自动打招呼脚本跑起来。覆盖两种用户路径——(A) 直接下载桌面 App（.app / .dmg / .exe）的非开发者，常见症状："Gatekeeper 拦"、"双击没反应"、".app 闪退"、"GUI 里配 API key"、"界面中英文怎么切"、"用不明白想让 AI 看看状态（「问 AI 帮忙」按钮）"；以及 (B) clone 源码 + uv 跑命令行的开发者，常见症状："uv 命令找不到"、"扫码扫不上"、"Chrome 起来空白"、"卡在哪一步"。当用户提到 BossZhiPin_Job_Search、BOSS 直聘脚本、自动招呼语、my_cover.pdf、chrome_profile、DEEPSEEK_API_KEY、Tauri、DMG 安装、PyTauri 等本项目特有概念，或者明显在按 README 走但被某一步卡住时，主动使用此 skill。即使用户没有明说"帮我装"，只要场景对得上就用。

Maintain a repo's Markdown docs (wiki chapters, ADRs, references) under Google's Open Knowledge Format (OKF), and find the right doc/schema fast. Two jobs. (1) Maintain docs the OKF way — each doc carries a small YAML frontmatter block as the single source of truth (type, title, tags, intent, schema_source, documents); schema detail links to the authoritative code instead of being copied into prose, so docs stop drifting and stop ballooning into thousand-line Markdown. (2) Look docs up fast — grep the literal term FIRST; only when grep is ambiguous (hits scattered across files / synonym mismatch / zero hits) run find_docs.py to rank the doc that owns the topic by frontmatter intent, or resolve a doc's schema_source straight to the code. Trigger phrases — "which doc covers X", "find the schema this doc points to", "where is endpoint/config-key X documented", "add OKF frontmatter to this doc", "lint the docs", "scaffold a new ADR/chapter". Run: scripts/run.sh find|schema|index|lint|new (or python3 scripts/find_

Deploy and configure MCBot, a Minecraft AI chat bot with RCON abilities. Use when the user wants to set up MCBot on a Minecraft server, configure AI providers (DeepSeek/OpenAI/Claude/Ollama), enable RCON, manage the bot service, or troubleshoot MCBot issues.

Take a still photo from the machine's local webcam with ffmpeg, then read/analyze or send the image. Cross-platform — Linux (V4L2), macOS (AVFoundation), Windows (DirectShow). Trigger when the user asks to "看看我", "take a photo", "take a picture", "selfie", "拍照", "自拍", "看看现在的环境/房间", or when the agent needs to visually observe the physical environment through the machine's camera. Requires ffmpeg and a local camera.

Take a still photo from a local V4L2 webcam (/dev/video0) with ffmpeg, then read/analyze or send the image. Trigger when the user asks to "take a photo", "take a picture", "selfie", "拍照", "自拍", "看看现在的环境/房间", or when Claude needs to visually observe the physical environment through the machine's camera. Linux-only; requires ffmpeg and a V4L2 device.