Skip to content

heyu-ai/yibi-stack

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

348 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

yibi-stack

CI Release Python License: Apache 2.0

English | 繁體中文


English

What is yibi-stack?

yibi-stack is a Claude Code skill and plugin stack for engineering teams who use Claude as their primary development tool. It transforms Claude Code from a capable-but-undisciplined assistant into a workflow-aware engineering partner with built-in guardrails.

The stack ships in two layers:

  • Plugins (plugins/) — installable via claude plugin install, add hooks and skills that activate automatically in every session
  • Skills (skills/) — SKILL.md runbooks that tell Claude how to approach specific tasks (TDD, PR review, spec writing, release workflow)

Why yibi-stack?

Without guardrails, Claude Code tends to:

  • Write bash commands that trigger Claude Code's internal confirmation dialogs (nested quotes, stateful cd, inline multi-language)
  • Generate specs too shallow to build from
  • Lose work context when conversations compress
  • Skip test discipline under time pressure

yibi-stack layers three forms of enforcement on top of Claude Code:

  1. Automated hooks — catch problems before execution, not after
  2. Methodology skills — runbooks that embed TDD, spec-writing, and review discipline into Claude's workflow
  3. Persistent tooling — mycelium, scheduler, and port registry that survive across conversations

Benefits

Benefit Mechanism
Safer shell execution bash-hygiene plugin intercepts 3 anti-pattern categories (AP1: overcomplex single commands / AP2: unicode in shell strings / AP3: stateful cd) via PreToolUse hooks — before they cause silent failures
Spec-first development sdd plugin + spectra-amplifier skill expands thin requirements into 5-layer specs (Context → Behavior → Data → Constraints → Acceptance Criteria), compatible with OpenSpec change management
Multi-model PR review pr-cycle-deep orchestrates Claude + Codex + Gemini in parallel independent review → cross-model debate → aggregate, catching issues no single model would flag
Persistent work memory mycelium skill auto-handovers before context compression and restores on next session start — no more losing track of multi-day work
Release discipline bump-version, protect-push, and ci-triage skills codify release workflow so Claude doesn't push to main without explicit intent
Test methodology tdd-kentbeck (Kent Beck Red→Green→Refactor) and qa-test-design (6 test design techniques) embed testing discipline into daily work

Architecture

plugins/          Claude Code plugin packs (installable via claude plugin install)
  bash-hygiene/   PreToolUse hook enforcement, shell hygiene rules, anti-pattern guide
  sdd/            Spec-Driven Development: Spectra + OpenSpec methodology, qa-test-design
  growth/         Session continuity: mycelium, learn, handover/newjob commands
  pr-flow/        PR workflow: review cycles, retrospective, bump-version, pr commands
  3rd-tools/      Third-party AI: Codex, Antigravity CLI (agy/Gemini), model verification
  tdd/            Test-Driven Development: Kent Beck TDD, Flutter TDD, CI triage
  util/           Utility: local port manager, debug command
  writing/        Writing quality: AI slop detection
  harness/        Harness readiness evaluation (README only, no package.json -- skill installed via make install)

skills/           Agent execution layer -- SKILL.md runbooks (installed via make install)
  <skill-name>/   Each skill is a flat directory with a SKILL.md runbook (or a symlink)
                  scope:global  -- works in any repo (methodology, cross-project tools)
                  scope:project -- requires this repo's Python tasks

tasks/            Python implementation (CLI, service, models, SQLite DB)
commands/         Claude Code slash commands (symlinked to ~/.claude/commands/)
scripts/          CI and lint tooling

Plugins vs Skills — what's the difference?

Plugins (plugins/bash-hygiene, plugins/sdd, plugins/growth, plugins/pr-flow, plugins/3rd-tools, plugins/tdd, plugins/util, plugins/writing) are proper Claude Code plugins with package.json manifests. They install hooks, rules, and bundled skills. Installable via claude plugin install without cloning. plugins/harness is a documentation placeholder only (no package.json) — its harness-eval skill installs via make install-one SKILL=harness-eval (see Plugins table).

Skills (skills/*/SKILL.md) are runbook files — not plugins. They're installed as symlinks into ~/.claude/skills/ via make install. They tell Claude how to approach a workflow; no hooks are involved. Skills are not individually installable via claude plugin install.

Install

Plugin-only (lightweight, no clone needed):

claude plugin marketplace add howie/yibi-stack
claude plugin install bash-hygiene@yibi-stack
claude plugin install sdd@yibi-stack
claude plugin install growth@yibi-stack
claude plugin install pr-flow@yibi-stack
claude plugin install tdd@yibi-stack
claude plugin install writing@yibi-stack

Note: harness-eval is not installable via claude plugin install — clone the repo and run make install-one SKILL=harness-eval (or make install-all for a full setup).

Full install (plugins + all skills + hooks + scheduler):

# 1. Install plugins (pre-execution hooks + rules)
claude plugin marketplace add howie/yibi-stack
claude plugin install bash-hygiene@yibi-stack sdd@yibi-stack growth@yibi-stack pr-flow@yibi-stack tdd@yibi-stack writing@yibi-stack

# 2. Clone and install skills + hooks + scheduler
git clone https://github.com/howie/yibi-stack
cd yibi-stack
make install-all

Verify install status (shows only this repo's skills, excludes gstack/external):

make status-own

Key Skills

Skill What it does
spectra-amplifier 5-layer spec expansion via sdd plugin
pr-review-cycle Full PR lifecycle: create -> parallel review (4 subagents) -> fix -> CI -> merge
pr-cycle-fast Fast lifecycle orchestrator: Python state machine, 1 reviewer, resumable
pr-cycle-deep Deep lifecycle: mob review (Claude + Codex + Gemini) + SDD amplifier-verifier
bash-anti-patterns AP1/AP2/AP3 detection guide + shell quoting hygiene reference
tdd-kentbeck Kent Beck TDD + Tidy First methodology
qa-test-design 6 test design techniques (equivalence, boundary, decision table...)
mycelium Cross-session work handover and insight collection
bump-version Version bump (Flutter/Python/Node/Go) + CHANGELOG + git tag
protect-push Hook to prevent accidental pushes from worktree branches to main
ci-triage CI failure triage funnel (Lint -> Type -> Security -> Tests)
learn Browse, search, prune, and export lessons learned
pr-retrospective 5-question PR retro, routes lessons to .claude/rules/ or CLAUDE.md
claude-md-prune Prune CLAUDE.md bloat: route gotchas to .claude/rules/ files, stay under 200-line soft cap
agy Antigravity CLI (Gemini) second opinion: lightweight single-model review / challenge mode
detect-ai-slop Detect AI-generated text patterns with model-specific fingerprints and de-slop suggestions
harness-eval 8-dimension harness readiness score (0-100) with PASS/WARN/FAIL checklist and priority TODO

See skills/README.md for the full index.

Plugins

Plugin Install Description
bash-hygiene claude plugin install bash-hygiene@yibi-stack Pre-execution bash anti-pattern detection with auto-fix guidance
sdd claude plugin install sdd@yibi-stack Spectra + OpenSpec spec-amplifier methodology + qa-test-design
growth claude plugin install growth@yibi-stack Session continuity: mycelium, learn, handover/newjob commands
pr-flow claude plugin install pr-flow@yibi-stack Full PR workflow: 5 skills + 6 slash commands
3rd-tools claude plugin install 3rd-tools@yibi-stack Codex, Antigravity CLI (agy/Gemini), model verification
tdd claude plugin install tdd@yibi-stack Kent Beck TDD, Flutter TDD, CI triage
util claude plugin install util@yibi-stack Local port manager + debug command
writing claude plugin install writing@yibi-stack AI slop detection
harness make install-one SKILL=harness-eval Harness readiness evaluation

繁體中文

yibi-stack 是什麼?

yibi-stack 是一套專為以 Claude Code 作為主力開發工具的工程師設計的 skill 與 plugin 集。它讓 Claude Code 從一個能力強但缺乏紀律的助手,升級為有工作流程意識、有自我約束力的工程夥伴。

這個 stack 分兩層:

  • Pluginsplugins/)— 透過 claude plugin install 安裝,每次 session 自動啟用 hook 與 skill
  • Skillsskills/)— SKILL.md runbook,告訴 Claude 如何處理特定任務(TDD、PR review、規格撰寫、發版流程)

為什麼需要 yibi-stack?

沒有護欄的 Claude Code 容易出現這些問題:

  • 寫出觸發 Claude Code 確認框的 bash 指令(巢狀引號、stateful cd、內嵌多語言)
  • 產生粒度太粗、無法直接實作的規格
  • 對話壓縮後遺失工作脈絡
  • 在時間壓力下跳過測試紀律

yibi-stack 在 Claude Code 之上疊加三層約束:

  1. 自動化 hook — 在執行前攔截問題,而不是事後除錯
  2. 方法論 skill — 把 TDD、規格撰寫、PR 審閱的紀律嵌入 Claude 的工作流程
  3. 持久化工具 — 跨對話的 session 記憶、定期排程器、port 登記表

主要好處

好處 機制
更安全的 shell 執行 bash-hygiene plugin 透過 PreToolUse hook,在執行前攔截三類反模式(AP1 過複雜單行 / AP2 bash 字串 Unicode / AP3 stateful cd),避免靜默失敗
規格先行的開發 sdd plugin + spectra-amplifier skill 把薄需求展開為五層規格(情境→行為→資料→約束→驗收條件),兼容 OpenSpec 變更管理框架
多模型 PR 審閱 pr-cycle-deep 讓 Claude + Codex + Gemini 並行獨立審閱再交叉辯論,捕捉單一模型漏掉的問題
持久化工作記憶 mycelium skill 在對話壓縮前自動交班,下次 session 開啟時自動恢復工作上下文,多日開發不斷線
發版紀律 bump-version + protect-push + ci-triage 讓 Claude 不會在沒有明確意圖的情況下推上 main
測試方法論 tdd-kentbeck(Kent Beck Red→Green→Refactor)和 qa-test-design(六大測試設計技術)把測試紀律內建到日常工作中

架構

plugins/          Claude Code plugin packs(可透過 claude plugin install 安裝)
  bash-hygiene/   PreToolUse hook 防線、shell 衛生規則、反模式修法指南
  sdd/            Spec-Driven Development:Spectra + OpenSpec 方法論、qa-test-design
  growth/         跨 session 連續性:mycelium、learn、handover/newjob commands
  pr-flow/        PR 全流程:review cycles、retrospective、bump-version、PR commands
  3rd-tools/      第三方 AI:Codex、Antigravity CLI(agy/Gemini)、模型驗證
  tdd/            測試驅動開發:Kent Beck TDD、Flutter TDD、CI 診斷
  util/           工具:local port manager、debug command
  writing/        寫作品質:AI slop 偵測
  harness/        Harness 就緒度評量(純 README,無 package.json,skill 透過 make install 安裝)

skills/           Agent 執行介面層(SKILL.md runbook,透過 make install 安裝)
  <skill-name>/   每個 skill 是一個目錄(或指向 plugins/ 的 symlink)
                  scope:global  -- 跨專案可用(方法論、通用工具)
                  scope:project -- 本 repo 限定(需要 tasks/ Python 實作)

tasks/            Python 實作(CLI、service、models、SQLite DB)
commands/         Claude Code slash commands(symlink 到 ~/.claude/commands/)
scripts/          CI 與 lint 工具腳本

Plugin 與 Skill 的差別?

Pluginplugins/bash-hygieneplugins/sddplugins/growthplugins/pr-flowplugins/3rd-toolsplugins/tddplugins/utilplugins/writing) 是有 package.json manifest 的正式 Claude Code plugin,會安裝 hook、rules 和隨附 skill,不需 clone 即可用 claude plugin install 安裝。 plugins/harness 是純文件容器(無 package.json)——其 harness-eval skill 透過 make install-one SKILL=harness-eval 安裝(見下方 Plugins 表格)。

Skillskills/*/SKILL.md)是 runbook 檔案,不是 plugin。透過 make install 以 symlink 安裝到 ~/.claude/skills/,告訴 Claude 如何執行特定工作流程。Skills 無法透過 claude plugin install 個別安裝。

安裝

只安裝 Plugin(輕量,不需 clone):

claude plugin marketplace add howie/yibi-stack
claude plugin install bash-hygiene@yibi-stack
claude plugin install sdd@yibi-stack
claude plugin install growth@yibi-stack
claude plugin install pr-flow@yibi-stack
claude plugin install tdd@yibi-stack
claude plugin install writing@yibi-stack

注意:harness-eval 無法透過 claude plugin install 安裝——需 clone 本 repo 後執行 make install-one SKILL=harness-eval(或 make install-all 一次裝齊)。

完整安裝(plugin + 所有 skill + hook + scheduler):

# 1. 安裝 plugin(pre-execution hook + 規則)
claude plugin marketplace add howie/yibi-stack
claude plugin install bash-hygiene@yibi-stack sdd@yibi-stack growth@yibi-stack pr-flow@yibi-stack tdd@yibi-stack writing@yibi-stack

# 2. Clone 並安裝 skill + hook + scheduler
git clone https://github.com/howie/yibi-stack
cd yibi-stack
make install-all

確認安裝狀態(只顯示本 repo 的 skill,排除 gstack / 外部安裝):

make status-own

主要 Skills

Skill 功能
spectra-amplifier 五層規格展開(透過 sdd plugin)
pr-review-cycle 完整 PR 生命週期:建立 → 並行 review(4 subagent)→ 修正 → CI → merge
pr-cycle-fast 快速 lifecycle orchestrator:Python state machine,可中斷 resume
pr-cycle-deep 深度 lifecycle:mob review(Claude + Codex + Gemini)+ SDD amplifier-verifier
bash-anti-patterns AP1/AP2/AP3 偵測指南 + shell 引號衛生參考
tdd-kentbeck Kent Beck TDD + Tidy First 方法論
qa-test-design 六大測試設計技術(等價類別、邊界值、決策表……)
mycelium 跨對話工作交班與洞察收集
bump-version 版本 bump(Flutter/Python/Node/Go)+ CHANGELOG + git tag
protect-push 防止 worktree branch 意外推上 main 的 hook
ci-triage CI 失敗快速診斷漏斗(Lint → Type → Security → Tests)
learn 瀏覽、搜尋、修剪、匯出教訓記錄
pr-retrospective PR 收尾五問回顧,路由 lesson 到 .claude/rules/ 或 CLAUDE.md
claude-md-prune CLAUDE.md 精簡:把 gotcha 路由到 .claude/rules/ 子檔,維持 200 行軟上限
agy Antigravity CLI(Gemini)第二意見:輕量單一模型 review / 對抗模式
detect-ai-slop 偵測 AI 生成文字特徵,含模型指紋比對與去 AI 味建議
harness-eval 8 維度 harness 就緒度評分(0-100),附 PASS/WARN/FAIL 清單與優先改善 TODO

完整索引見 skills/README.md

Plugins

Plugin 安裝指令 說明
bash-hygiene claude plugin install bash-hygiene@yibi-stack 執行前 bash 反模式偵測,附自動修法指引
sdd claude plugin install sdd@yibi-stack Spectra + OpenSpec 規格展開 + qa-test-design 測試設計
growth claude plugin install growth@yibi-stack 跨 session 記憶:mycelium、learn、handover/newjob
pr-flow claude plugin install pr-flow@yibi-stack 完整 PR 流程:5 個 skill + 6 個 slash command
3rd-tools claude plugin install 3rd-tools@yibi-stack Codex、Antigravity CLI(agy/Gemini)、模型驗證
tdd claude plugin install tdd@yibi-stack Kent Beck TDD、Flutter TDD、CI 診斷
util claude plugin install util@yibi-stack 本機 port 管理 + debug command
writing claude plugin install writing@yibi-stack AI slop 偵測
harness make install-one SKILL=harness-eval Harness 就緒度評量

License

MIT

About

Agentic skill stack for Claude Code: bash hygiene, Spectra/OpenSpec methodology, PR review workflows, TDD, and productivity tools.

Resources

License

Stars

Watchers

Forks

Packages

 
 
 

Contributors

Languages