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 hcom-spawngit 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/hcom-spawn)<a href="https://agentmods.dev/skills/ouob-tw/loopagentteams/hcom-spawn"><img src="https://agentmods.dev/badge/skills/ouob-tw/loopagentteams/hcom-spawn/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/hcom-spawn"><img src="https://agentmods.dev/badge/skills/ouob-tw/loopagentteams/hcom-spawn.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 | $0.00080 | $0.03676 |
| Opus 5.5 | $0.00032 | $0.01470 |
| Sonnet 5.5 | $0.00016 | $0.00735 |
| Haiku 4.5 | $0.00008 | $0.00368 |
Grade A, and why
hcom-spawn 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 3d 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.
透過 HCOM 召喚 Agent
你是召喚者：選模型、啟動、交代任務、追蹤存活、向使用者回報、收尾。被召喚的 Agent 彼此直接溝通，你不逐句轉達。
需要獨立 worktree、多人同時改同一批檔案，或要完整的討論→Spec→Tickets→QA 流程時，改呼叫 /lat。
選模型與 effort
使用者指定的模型與 effort 優先。使用者只說「Codex」或「Claude」時，用該 client 的預設模型。未指定時：前端任務用 Claude，簡單的用 claude-sonnet-5-5、複雜的用 claude-opus-5-5；後端任務用 Codex（gpt-6.1-sol，簡單複雜都從 medium 起）；其他或混合任務依任務內容選。
| 模型 | 起始 effort |
|---|---|
gpt-6.1-sol（後端預設） |
medium |
gpt-6-astra |
簡單任務 low、複雜任務 medium |
gpt-5.6-sol、claude-opus-5-5、claude-sonnet-5-5、其他 |
medium |
簡單任務：查資料、讀文件或程式碼回答問題、跑指令收集結果、單一檔案小修改、照明確指示改文字。 複雜任務：跨檔案實作、除錯、設計、審查、需自行權衡方案。
上表是常用預設，不是白名單。使用者指定表外模型時，先用工具查該 client 的可用模型，必要時查官方文件，確認正確 ID、可用性與支援的 effort。能唯一對應就用；有歧義或無法使用才詢問，不擅自換模型。表外模型從 medium 起；不支援時用已查證的可用級別並說明。
啟動時明確傳入模型 ID 與 effort，不依賴隱含預設。前後端分工只看任務，不因為目前 client 的內建 Agent 做不到就改分工。
用自己 client 的內建 Agent 時
模型與 effort 照上面同一套規則，用 client 支援的欄位傳入。權限受宿主限制，把 CLI 參數寫進 prompt 不等於生效。內建機制套不上必要設定時，改用支援的 HCOM 外部 Agent；仍不可用就回報使用者。
啟動
tag 用 <自身名稱>-<主題>，主題一兩個英文小寫單字（review、docs、api-probe）。HCOM 會把 Agent 名稱變成 <tag>-xxxx，所以同一位召喚者開的每組都認得出來，也能用 @<tag>- 一次傳給整組。
hcom 1 codex --tag <自身名稱>-<主題> --go --name <自身名稱> \
--model gpt-6.1-sol -c 'model_reasoning_effort="medium"' --yolo
hcom 1 claude --tag <自身名稱>-<主題> --go --name <自身名稱> \
--model claude-opus-5-5 --effort medium --dangerously-skip-permissions
Codex 預設加 --yolo，Claude 預設加 --dangerously-skip-permissions，讓被召喚的 Agent 不會停在確認畫面卡住協作。--yolo 同時關掉確認與 sandbox。這些啟動參數不擴大任務授權範圍，任務邊界寫在第一則訊息裡。
使用者指定 HERDR workspace 時，先 herdr workspace list 查目標 ID，再 HERDR_WORKSPACE_ID=<目標ID> hcom 1 <client> --terminal herdr ...。未指定就照常執行，Agent 會開在呼叫端所在的 workspace。
啟動後核對
啟動輸出的 Names: 只印 base name，完整名稱是 <tag>-<base>，用 hcom list --name <自身名稱> 取。Claude 有時超過 10 秒的就緒等待而回報 Still launching，那不是失敗，重查 hcom list 即可。
hcom term <名稱> --name <自身名稱> 看畫面核對三項：
| Client | 模型與 effort | 權限 |
|---|---|---|
| Codex | 底部狀態列 <模型名> <effort>（如 GPT-6.1-Sol medium） |
開場 permissions: YOLO mode |
| Claude Code | 開場 Opus 5.5 with <effort> effort、底部 Opus 5.5 <effort> |
底部 ⏵⏵ bypass permissions on |
What ships with it
1 file 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.
- 3d ago First seen · 168 lines · 80 tokens per session scan A e43bfff12864
hcom-spawn is a skill published in the GitHub repository ouob-tw/LoopAgentTeams (14 stars, last pushed today), licensed MIT. It adds 80 tokens to every session and 3,676 once invoked, about $0.0003 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
loki-mode
An autonomous software factory that knows what it is supposed to deliver, and proves it did. Use when the user says Loki Mode or asks to build, fix or verify software autonomously.
checkpoint-mode
Pause for review every N tasks - selective autonomy pattern.
loki-mode
Launch Loki Mode autonomous SDLC agent. Handles PRD-to-deployment with minimal human intervention. Invoke for multi-phase development tasks, bug fixing campaigns, or full product builds.
prompt-optimization
Applies prompt repetition to improve accuracy for non-reasoning LLMs.
loki-seal
Use when finishing any coding task in a repo with tests. Loki Seal runs the repo's real test suite and refuses to let you declare done while tests newly fail, or while tests or CI config were deleted, skipped, xfailed or weakened. Runs locally with no model calls.
engine-comments
Decide where a piece of reasoning belongs when writing or editing this engine — an invariant beside the code, a transition in the commit message, or a decision in decisions/. Use when adding a comment or a file header to hooks/, scripts/, commands/ or agents/, when a header has grown past what a reader needs, or when…