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 skills/tienenwu/fables/androidnpx skills add tienenwu/fables --skill androidgit clone --depth 1 https://github.com/tienenwu/fablesWhat 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 | $0.00080 | $0.01239 |
| Opus 5 | $0.00040 | $0.00620 |
| Sonnet 5 | $0.00016 | $0.00248 |
| Haiku 4.5 | $0.00008 | $0.00124 |
Grade A, and why
android-dev-playbook 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 2d ago.
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 — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.
🌐 English version · 繁體中文（正本 / canonical）
Android 開發判準手冊
核心原則
- Debug build 的綠燈不算數：R8、混淆、簽章、後端環境旗標都只在 release 生效——release 路徑的改動必須用 release 級驗證。
- 狀態只往下流、事件只往上傳：UI 層不做業務判斷，Composable 不持有可變業務狀態。
- 生命週期是預設敵人：任何持有 Context/View 的長生命週期物件，先假設它會洩漏，再證明不會。
- 反射是 R8 的盲區：所有經反射建構的類別（Gson/Moshi/Retrofit model）改動時，keep 規則同步改，否則是 release-only 炸彈。
- 主執行緒上不做 IO，協程不用 GlobalScope——沒有例外情境值得討論。
開工分流
| 情境 | 路徑 | 先讀 |
|---|---|---|
| 新增畫面/功能 | 先定狀態擁有者與資料流，再寫 UI | references/architecture-state.md |
| UI 卡頓、recomposition 過多 | 不要先改 UI，先查狀態設計 | references/compose-ui.md |
| 非同步、Flow、生命週期問題 | 查 collect 位置與 scope 選擇 | references/coroutines-lifecycle.md |
| 加/改 API model（Gson/Moshi/Retrofit） | 實作 + keep 規則一起改 | references/release-checklist.md §R8 |
| 出 release / 改 build 設定 | 逐條跑必查清單 | references/release-checklist.md |
| release 才爆、debug 正常 | 先假設 minify/混淆/環境旗標，別亂猜 | references/release-checklist.md |
紅線（絕對禁止）
- 禁止只用 debug build 驗證後宣告 release 相關改動完成——minify、keep 規則、BuildConfig 旗標在 debug 全部不生效。
- 禁止
GlobalScope.launch——生命週期不受控，畫面銷毀後繼續跑，洩漏加 crash。 - 禁止在 Composable 函式本體做網路/DB 呼叫——每次 recomposition 重打一次，用
LaunchedEffect或下放 ViewModel。 - 禁止 ViewModel 持有 Activity/Fragment/View 的參照——旋轉螢幕即洩漏；要 Context 用 Application 級。
- 禁止 catch 後只 log 不通知呼叫端——使用者會看到永遠轉圈的 spinner；錯誤要變成 UI state。
- 禁止改測試斷言讓 CI 變綠——連紅兩次是方向錯誤訊號，退回上一決策點。
失敗訊號（該回頭，不是重試）
| 徵兆 | 多半是 | 退回 |
|---|---|---|
| 修 recomposition 一處、另一處又卡 | state 放錯層級，粒度太粗 | 重看狀態擁有者設計 |
remember / derivedStateOf 越加越多才不閃爍 |
上游把不穩定物件當 state 傳 | 檢查傳入參數的穩定性 |
| release crash 修一個 keep 規則又冒下一個 | 逐類補洞，該整包 keep 該 model 目錄 | R8 規則層 |
| 生命週期 callback 裡的 null check 越寫越多 | 持有了不該持有的參照 | 所有權設計 |
| 同一個 ANR 換三種寫法還在 | 主執行緒上有隱藏的同步 IO | 找出真正阻塞點再動手 |
references 索引
references/architecture-state.md— 分層判準、狀態該放哪一層、UDF 正反例。動架構或新功能前讀。references/compose-ui.md— 何時抽 Composable、穩定性與 recomposition、副作用 API 選擇。寫 UI 前讀。references/coroutines-lifecycle.md— scope/Dispatcher 選擇、Flow collect 位置、洩漏典型。碰非同步就讀。references/release-checklist.md— R8/keep、build variant、簽章、權限、出版前逐條打勾。出 release 必讀。references/test-scenarios.md— 判準測驗集，驗證接手模型是否照走。不給執行中的模型讀。
What ships with it
5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 2d ago First seen · 55 lines · 80 tokens per session scan A 87b26ce4f57b
android-dev-playbook is a skill published in the GitHub repository tienenwu/fables (4 stars, last pushed 1mo ago), licensed MIT. It adds 80 tokens to every session and 1,239 once invoked, about $0.0004 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 skills, from other repositories
find-release
A skill to find the lowest Dart and Flutter release containing a given commit. Use this skill whenever users ask about when a commit landed in Flutter or Dart releases, inquire about release versions for specific SHAs, or want to know if a commit is included in stable, beta, or dev channels for Flutter/Dart projects.
speckit-specify
Create or update the feature specification from a natural language feature description.
speckit-analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
docs-build
Build, preview, and validate the Uno documentation website (DocFX) locally — surface real content errors vs expected noise, drive rendered pages with Playwright, and validate external-doc commit bumps in importexternaldocs.ps1 before a PR. Use when working under doc/, editing articles/ markdown, bumping an external…
speckit-taskstoissues
Convert existing tasks into actionable, dependency-ordered GitHub issues for the feature based on available design artifacts.
speckit-git-feature
Create a feature branch with sequential or timestamp numbering.