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 ouob-tw/LoopAgentTeams --skill latgit clone --depth 1 https://github.com/ouob-tw/LoopAgentTeamsWrote 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/ouob-tw/loopagentteams/lat)<a href="https://agentmods.dev/skills/ouob-tw/loopagentteams/lat"><img src="https://agentmods.dev/badge/skills/ouob-tw/loopagentteams/lat/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/ouob-tw/loopagentteams/lat"><img src="https://agentmods.dev/badge/skills/ouob-tw/loopagentteams/lat.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 | $0.00038 | $0.04282 |
| Opus 5.5 | $0.00015 | $0.01713 |
| Sonnet 5.5 | $0.00008 | $0.00856 |
| Haiku 4.5 | $0.00004 | $0.00428 |
Grade A, and why
lat 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 22d 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 — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.
LoopAgentTeams
LAT 管理角色、階段轉換與交接；各階段的做法交給對應 Matt skill。
啟動時讀取專案 AGENTS.md／CLAUDE.md 與 docs/agents/issue-tracker.md（由 /setup-matt-pocock-skills 設定），依配置的 issue tracker 讀寫 Spec 與 Tickets。尚未設定時，若專案 repo 是公開的，建議另建私人 <repo>-work repo 存放 issues，避免交接與證據中的內網位址、本機路徑或帳號資訊公開。證據、交接／未解事項及學習說明分開保存，具體位置依專案設定；交接與未解事項建立 tracker issues，決策依下方「交接」規則保存。
Codex 主控恢復
Codex 主控真正啟動 LAT 時，讀完專案設定後，第一個動作依 Codex 恢復程序 準備既有進度索引並 activate，再建立執行清單；只閱讀／審查技能或擔任執行 Agent 不啟用。每個檢查點更新索引。compact／resume 收到提示時，依該程序完整重讀規則、references、進度與待決紀錄，核對 tracker 及授權後續作；交付／取消時 deactivate。Claude 維持原流程。
角色
- Orchestrator：主控 Agent，管理流程、派發、裁決與整合交付；詳細執行交給其他 Agent。
- Designer：與使用者討論，接續整理 Spec 與 Tickets；可由主控兼任或外部 Agent 擔任。
- Implementer：實作、測試、呼叫獨立 Review、修正與提交。
- QA：操作真實應用，依需求與驗收條件測試；與實作者分開。
執行清單
Orchestrator 啟動時，將下列清單建立為本次任務的進度紀錄。收到委派結果後，核對完成條件再勾選，並推進下一個可執行步驟。進度記錄在任務中，SKILL.md 保持為模板。
- 討論：Designer 使用 /grill-with-docs 與使用者釐清需求，確認方向。
- Spec：同一位 Designer 使用 /to-spec，整理需求、範圍與驗收條件。
- 使用者確認：讓使用者確認 Spec；使用者已明確確認目前版本的 Spec 時，不重複詢問。這是唯一的正式確認點。
- Tickets：同一位 Designer 使用 /to-tickets，拆分工作並列出驗收條件與依賴。
- 同步派 QA 依下方「測試清單」從使用者需求獨立列出情境，作為實作前草稿。
- 派發：Orchestrator 核對規劃與分工。
- 若本票只完成已確認需求的一部分，其餘必要工作須同時建立 tracker Ticket，列明驗收條件與依賴，不只寫「之後補做」。超出已確認需求的功能，不自行加票。
- 依 任務卡 建立每張派工卡，配置各 Agent 的獨立 sandbox／container prefix 與 port range，登記資源歸屬。
- 平行派發前，列出各 Ticket 預計修改的檔案與整合後必須成立的限制，先協調解決衝突。
- 共用檔案依序修改，指定目前負責的 Ticket；接手前由 Orchestrator 更新分工與依賴，下一張 Ticket 從已整合版本接續。
- 將可開始的 Tickets 交給 Implementer，每張實作任務使用新 context。
- 實作：Implementer 使用 /implement；本輪派發的 Tickets 完成實作、測試、獨立 Review 與必要修正。
- Implementer 依下方「測試清單」提供實際實作的路徑清單，Reviewer 在既有 Review 中核對遺漏，修正後更新。
- Orchestrator 自行實作的部分同樣送獨立 Review，標準與其他 Tickets 相同； 兼任實作者不等於免除審查，未經審查的那一段通常是整批品質最差的地方。
- 整合：Orchestrator 協調整合，由執行 Agent 在整合後版本驗證各 Ticket 的驗收條件與跨 Ticket 限制。
- 合併到整合分支前先公告；從 Git 核對交付 commit 的存在、內容與整合結果，從實際測試與有效證據核對成果，不直接採信 Agent 回報的 hash 或成功敘述。
- 交付版本須保留原樣的檔案，記錄路徑與基準 commit，逐一做位元組比對，合併無衝突時亦同；「本票不修改」不代表其他 Ticket 不得修改。
- 驗證在整合後工作區執行，記錄版本、環境重建方式、指令與結果；指向其他 Ticket 工作區的絕對路徑須改為整合後工作區的對應路徑再執行。
- 同版本且環境與設定仍適用的有效證據可沿用，其餘重新驗證。
- 每項任務合併至整合分支且整合後驗證通過後，立即依 任務收尾 關閉 Agent、清理及歸檔，不等整批結束；再重新計算可開始的 Tickets。 新解除阻塞的項目重新進入「派發」，含其下的分工與限制協調。所有 Tickets 完成前，流程不停在階段交接點。
- 驗收：Orchestrator 與 QA 先依下方「測試清單」合併需求與整合後實作的情境，再逐項執行；問題交回 Implementer 修正，更新受影響項目後複驗。
- 使用者介面的驗收條件，須透過真實 UI 按使用者操作方式驗證；直接呼叫 API 可用於準備資料或診斷，不算 UI 驗收通過。
- 對先前未攔下的缺陷，依證據辨別原因：缺少情境、走錯路徑、斷言不足、未執行／環境不符，或測試替換行為掩蓋缺陷；未知則保留未知。在同一修正與複驗流程補足覆蓋，回歸證據依 three-tier-testing 的「測試可信度」要求處理。
- 交付：Orchestrator 核對驗收證據，確認測試資源已依 three-tier-testing 清理或交接，並核對各任務已完成收尾或在任務卡明列待完成後續工作。回報成果與未完成事項；必要驗證未執行時，不宣告完成。
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.
- 22d ago First seen · 111 lines · 38 tokens per session scan A 86a1fc3fc5a8
lat is a skill published in the GitHub repository ouob-tw/LoopAgentTeams (20 stars, last pushed yesterday), licensed MIT. It adds 38 tokens to every session and 4,282 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-10-01.
Other skills, from other repositories
dispatch
Use when a task file exists in .hyperflow/tasks/ and workers need dispatching. Fans out parallel workers under per-batch Reviewers, runs a final integration review, and commits per sub-task. Endpoint of the auto-chain — no auto-deploy. Trigger with /hyperflow:dispatch, "run the plan", "execute the task", "build it"…
gsd-new-project
Initialize a new project with deep context gathering and PROJECT.md.
gsd-ns-manage
Invoke this Pi skill with /skill:gsd-ns-manage.
gsd-resume-work
Resume work from previous session with full context restoration.
spec-kitty-mission-system
Understand how Spec Kitty missions work: the 4 built-in mission types, how they define workflows via step contracts and action indices, how missions and work packages relate, how templates are resolved through the 6-tier chain, and how doctrine artifacts (procedures, tactics, directives) compose mission behavior.…
spec-kitty-runtime-review
Review runtime-owned outputs using the Spec Kitty review workflow surface, then direct approval or rejection with structured feedback. Triggers: "review this work package", "check runtime output", "approve this step", "review WP", "is this WP ready to approve", "check this implementation". Does NOT handle: setup-only…