📍 Chengdu · 🤖 AI workflow builder · 📓 AI Notes by huiyonghkw
I run every workflow end-to-end before I show it — then open-source the Skills and ship the tooling.
Three self-hosted tools I actively maintain — start with the problem you want to solve:
| 🔐 CertHub | 🖥️ hekouwang-terminal-kit | 📝 hekouwang-typora-theme |
|---|---|---|
| Take SSL certificates from issuance to production: renew, SSH deploy, manual ZIP, full-chain verification. | Configure iTerm2, Ghostty, Warp, Terminal.app and the CLI toolchain from one palette — safe install and uninstall. | Bring the Claude desktop reading feel to Typora through a maintainable token-driven theme. |
| Product home → | Product home → | View repository → |
More open-source tooling: CLAUDE.md Doctor · Skill Doctor · Environment Doctor · Yandu Deck
- 🩺 hekouwang-claude-md-doctor-skill —
CLAUDE.mdlinter for harness repos - 🧪 hekouwang-claude-skill-doctor-skill —
SKILL.mdquality gate before you publish a Skill - 🧹 hekouwang-env-doctor-skill — dev-env doctor with interactive uninstall picker
- 📈 hekouwang-stock-data-reader-skill — public stock data via akshare → neutral recap (compliance guardrails built in)
- 🖥️ hekouwang-terminal-kit — featured product above
- 📝 hekouwang-typora-theme — Claude-desktop reading experience in Typora
- 🐍 hekouwang-laravel-to-fastapi — Laravel ↔ FastAPI side-by-side for muscle-memory transfer
- 🎬 hekouwang-yandu-deck-skill — Yandu Deck slide pipeline
- 🌐 theme.typora.io — Typora themes site
🔒 = private (404 unless you have access). The harness behind AI Notes by huiyonghkw.
- 📦 hekouwang-content-skill — one draft → finished assets for every platform (article / cards / video), six visual systems
- 🏭 hekouwang-content-harness — personal AI harness master: rules, recipes, multi-tool loading
- 🧠 hekouwang-content-memory — cross-session memory: pitfalls / contracts / preferences (
MEMORY.mdindex) - 📊 hekouwang-channel-analyzer — Xiaohongshu / Toutiao / WeChat analytics warehouse + Observable dashboard
- 🎨 hekouwang-xhs-theme-skill — Xiaohongshu card visual theme library
- 🚕 cczhaoche-harness — AI-collaboration harness for a ride-hailing SaaS (meta-layer only; business code on Aliyun Codeup)
- 🧠 cczhaoche-memory — cross-session memory for the same client
- 🎬 cczhaoche-product-anim-skill — product intro animations → Aliyun OSS
- 📱 cczhaoche-passenger-prototype-skill — passenger-app UI system (OKLCH + BEM)
- 🛣️ ridehailing-mcp — ministry-level ride-hailing reporting MCP (CaaS)
Legacy · pre-AI era (gitlab-ci-docker 302★ · lnmp-docker 237★)
- 🐳 gitlab-ci-docker · lnmp-docker — Docker CI/CD stacks from my backend years
- 🏪 zhanye / zhanye-h5 / zhanye-console — 有信展业 client / H5 / admin
- 🚗 api.chehubao.com — Chehubao API
- Running AI Notes by huiyonghkw — explain dense AI in plain talk, ship Skills readers can copy, record every workflow before I teach it
- Building the doctor suite — lint
CLAUDE.md,SKILL.md, and dev environments the way you'd lint code - Shipping terminal + harness tooling — one palette across four terminals; one harness across Claude Code, Cursor, and Codebuddy
- Writing on hekouwang.pages.dev — long-form articles, Yandu Decks, open recipes
- 📓 CN platforms — WeChat · Xiaohongshu · Toutiao · Bilibili: search 「会勇禾口王的AI笔记」
The question isn't whether AI will replace you — it's how the people who use AI first will. I don't teach from slides; I ship from my own machine, then hand you the Skill.
About the name · 禾口王 → 程 · 会勇 → 会用
「禾口王」 stacks into 「程」 (chéng, code / process); 「会勇」 sounds like 「会用」 ("knows how to use") — 会勇禾口王 = a programmer who genuinely knows how to use AI.


