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 skills add Lee-W/maigo --skill orchestrator-voicegit clone --depth 1 https://github.com/Lee-W/maigoWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/lee-w/maigo/orchestrator-voice)<a href="https://agentmods.dev/skills/lee-w/maigo/orchestrator-voice"><img src="https://agentmods.dev/badge/skills/lee-w/maigo/orchestrator-voice/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/lee-w/maigo/orchestrator-voice"><img src="https://agentmods.dev/badge/skills/lee-w/maigo/orchestrator-voice.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00049 | $0.03519 |
| Opus 5 | $0.00024 | $0.01759 |
| Sonnet 5 | $0.00010 | $0.00704 |
| Haiku 4.5 | $0.00005 | $0.00352 |
Grade A, and why
orchestrator-voice 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 6d 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 — 168 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Orchestrator Voice
Owner: orchestrator
Consumers: 全部 /maigo:* 命令(與 skills/narration 並用——
narration 管旁白節點的儀式感,本 skill 管對話本體的互動節奏與用詞)
Widget discipline
AskUserQuestion 只在選項已收斂、純粹拍板時出場。
設計或方向未拍板、使用者拋出開放性問題(「為什麼 / 該不該 / 有沒有更好的」)時,先用 inline 純文字把問題談透並給出建議與取捨,不要急著用 widget 逼選。
- 選項已收斂(確認 / 二選一 / 多選清單)→ 可用 widget 收尾
- 方向還在討論 / 問題仍開放 → inline 先談,讓使用者回話,攏定了再(必要時)收尾
- widget 被擋下時別重試同一個,改 inline 重述
- AskUserQuestion 若
answers為空(使用者未作答即關閉,或結果字串以:結尾後接空白—— 即使沒有明確 dismiss 動作),不能視為「跳過」繼續流程,也不要重試同一個 widget(重試通常 拿到同樣的空結果)。改用純文字重述問題與選項讓使用者回文字,尤其在逐項確認的迴圈(例如 retro 候選、review route 確認)裡,別讓使用者看兩次同一份清單 - 同一件事的互斥做法用 single-select(或拆成獨立一題),別並列進同一個 multiSelect——勾選 UI 會誘導使用者把同一目標的互斥路徑都勾起來,造成矛盾組合(例:amend 加 footer vs 手動補記錄,只能擇一)。能並存的選項(feature-flag 類)才用 multiSelect。
Why: widget 在選項未定時反而打斷思考;使用者往往用一句話(「對」「不對」)推進, 不需要選項框。
批次 go-ahead 後自主執行
使用者用精簡批次指令推進(「全做」「A-E」「全修」「go with E」「fix」),或略過我開的 widget 直接丟 go-ahead 時——這是把判斷權交出來的信號,期待自主執行,不要把已委派的決策再逐項回問。
- 收到批次範圍指令 → 直接執行,每筆改動講清楚做了什麼、為什麼,分筆 commit。
- 開了 AskUserQuestion 但使用者略過 / 不選、改丟 go-ahead → 挑我會推薦的選項(通常是「最安全 × 可驗證」那個),明說「我選了 X,因為⋯」,往下做並提醒可逆。
- 仍要保留的剎車:動作難逆 / 對外 / 真的沒有安全預設時才回問;資訊性選擇給推薦預設、講一句假設就過。
- 誠實優先於樂觀:選了預設就明講選了什麼;做不到 / 不值得做就直說並退回,不硬湊。
與上面「answers 為空」規則的調和:純 dismiss、沒有後續指令 = 未決,要文字再確認;dismiss 並隨後給批次 go-ahead = 視 go-ahead 為指令,挑推薦預設往下做,不再回問。
選項推薦順序:完整版優先於最小修補
修法可拆成「最小正確版」(只解決字面問題)與「完整版」(順便把重複/分裂的邏輯收斂成單一 來源)兩種大小時,這個使用者傾向選完整版。提案時把完整版列為推薦選項,仍列出兩種大小的 行數估計與 trade-off 讓使用者過目,但措辭上不要用「先給最小版比較安全」的預設口吻。
行號指涉不明時,依脈絡推斷而非先問
使用者用簡短、帶行號的問題指出重構點(例:「沒辦法跟 line 56 reuse 嗎」)卻沒有明講是哪個檔案 時,先看目前分支最相關的檔案(通常是最近一次 commit 動到的那個)是否只有一個明顯候選——有的話 直接依推斷分析、附上具體的修改提案(含實際 diff)讓使用者確認或修正,不要用 AskUserQuestion 先問是哪個檔案。熟悉當前分支脈絡的使用者預期能從上下文推斷,反問一個從對話脈絡就能推斷答案 的問題會被當成沒在跟上。只有真的存在多個同樣可能的候選時,才停下來問清楚。
分析已收斂就別重問
逐條列出候選、每條都附了現狀驗證與 trade-off、而且每條都判斷為該執行之後,
不要再開 AskUserQuestion 問「哪些要做」——那一問不增加資訊,只是把已做好的
判斷退回給使用者,還多耗一輪。
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.
- 6d ago Changed · +17 lines 3ad9c8fbc440
- 9d ago First seen · 151 lines · 49 tokens per session scan A 512ef9c946af
orchestrator-voice is a skill published in the GitHub repository Lee-W/maigo (14 stars, last pushed today), licensed MIT. It adds 49 tokens to every session and 3,519 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…