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 cjhyy/codeshell --skill loop-orchestrationgit clone --depth 1 https://github.com/cjhyy/codeshellWrote 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/cjhyy/codeshell/loop-orchestration)<a href="https://agentmods.dev/skills/cjhyy/codeshell/loop-orchestration"><img src="https://agentmods.dev/badge/skills/cjhyy/codeshell/loop-orchestration/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/cjhyy/codeshell/loop-orchestration"><img src="https://agentmods.dev/badge/skills/cjhyy/codeshell/loop-orchestration.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 | $0.00109 | $0.01081 |
| Opus 5.5 | $0.00044 | $0.00432 |
| Sonnet 5.5 | $0.00022 | $0.00216 |
| Haiku 4.5 | $0.00011 | $0.00108 |
Grade A, and why
loop-orchestration 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 — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Loop Orchestration(编排文档生成器)
产出一份 .loop/<slug>/ORCHESTRATION.md——定义一个 loop 里「谁干什么」。
它是 loop-mode 执行器的输入。本 skill 只生成/校验编排文档,不执行 loop。
什么时候用
- 用户要配置一个 loop 的角色分工 / 计划-执行流程时。
loop-mode找不到 ORCHESTRATION.md 时,回退到这里先生成。
两种来源
- 用户已写死:若
.loop/<slug>/ORCHESTRATION.md已存在 → 读它、校验字段齐全即可,不要重新生成。 - 访谈生成:否则问下面几个问题(一次问一组,别轰炸),据答案写出文档。
访谈要点(用 AskUser,尽量给选项)
- slug / 任务名:给这个 loop 起个短名(目录
.loop/<slug>/)。 - 角色→执行者:每个角色选
codex/claude-code/self(codeshell 自己直接干):planner出蓝图 / 拆 milestone(可选,纯打磨 loop 不需要)executor干活出提案(必填)reviewer审查提案(可选,默认 self)
- 阶段:
milestone(照蓝图推进)/polish(发散打磨)/milestone→polish(先推进后打磨)。 - 蓝图来源(若含 milestone 阶段):指一个现成文档路径 / 让 planner 现场拆。
- 值守:Attended(有人,高风险停等)/ Unattended(无人,高风险堆待批继续)。
- 闸门:额度/时间/轮数上限(你把关的停止条件)。
产出模板(写到 .loop//ORCHESTRATION.md)
# Orchestration: <slug>
## Goal
<一句话目标>
## Roles
- planner: <codex|claude-code|self|none>
- executor: <codex|claude-code|self>
- reviewer: <codex|claude-code|self>
## Phases
<milestone | polish | milestone→polish>
## Blueprint
- source: <文档路径 | "planner 现场生成" | none>
## Attendance
<attended | unattended>
- 切换点(milestone→polish): attended=问一次 / unattended=自动切
## Permissions(最小必要,按值守分级)
- 默认: acceptEdits
- 提权: 逐轮显式、有理由;unattended 禁网络写/远端写/凭证/仓库外写(除非 goal 显式授权)
- 不可逆交付永远留待批
## Gates
- <额度 / 时间 deadline / 最大轮数 / 单轮上限>
校验(生成或读取后都做一遍)
executor必填;planner=none时 Phases 不得含 milestone。- Phases 含 milestone 时 Blueprint.source 必须有值。
- 角色执行者只能是
codex/claude-code/self。 - 缺字段 → 补问用户,别猜。
候选分支与审查契约
文档还必须写明:
- candidate:
loop/<slug>/candidate;通过验证和 reviewer 才合入 candidate。 - proposal: 每轮
loop/<slug>/rNNN,从 candidate 分出。 - delivery: 不自动 push,不自动合入 main。
- ledger writer: 只有协调者可写
.loop/<slug>/STATE.json和RUNS.jsonl;执行者/审查者只返回回执。 - review independence: reviewer 默认与 executor 不同;高风险轮次不得 self-review。
交接
文档就绪后告诉用户:「编排已写入 .loop/<slug>/ORCHESTRATION.md,用 /loop <目标> 启动;要无人运行就把‘夜间’写进目标,无需打开 Goal。」不要在本 skill 里直接开始跑 loop。
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 First seen · 79 lines · 109 tokens per session scan A 40df8613ac75
loop-orchestration is a skill published in the GitHub repository cjhyy/codeshell (65 stars, last pushed 2d ago), licensed MIT. It adds 109 tokens to every session and 1,081 once invoked, about $0.0004 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 2026-10-14.
Other skills, from other repositories
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…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
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
Optimize the meaningful UI available immediately from a Next.js route on an initial load (hard navigation) or named client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route and entry point at a time; the shipped test then guards…