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/kangarooking/loop-engineering-skill/loop-5plus1-architecturenpx skills add kangarooking/loop-engineering-skill --skill loop-5plus1-architecturegit clone --depth 1 https://github.com/kangarooking/loop-engineering-skillWhat 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 | $0.00116 | $0.01550 |
| Opus 5 | $0.00058 | $0.00775 |
| Sonnet 5 | $0.00023 | $0.00310 |
| Haiku 4.5 | $0.00012 | $0.00155 |
Grade A, and why
loop-5plus1-architecture 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 — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Source Metadata
Original cangjie-skill frontmatter from the distillation run:
name: loop-5plus1-architecture
description: |
设计或审计完整 Loop 系统的组件架构: 5 大组件 + 1 根脊柱。
当用户需要搭建一个完整的 (而非最小可用的) loop 系统、或审计现有系统的组件完整性时使用。
不适用于: 最小可用 loop 的设计、或单次任务的自动化。
关键 trigger: "设计一个完整的 loop 系统"、"我的 loop 缺什么"、"loop 系统怎么组织"。
source_book: "Loop Engineering 视频合集"
source_chapter: 视频3 (小木头) / 视频4 (Idoos Money)
tags: [architecture, system-design, components, blueprint]
related_skills: [loop-three-elements, maker-checker, goal-verification]
5+1 循环系统架构 — 完整 Loop 的设计蓝图
R — Reading (原文)
"ados money把它拆成了5块,外加一根脊柱。第一块心跳(定时触发),第二块工作树(work tree,隔离分支),第三块skill(项目规则),第四块连接器(MCP,接入真实工具),第五块子智能体(写代码和审代码分开)。最后就是那根脊柱——支撑起整个loop循环系统的记忆。" — 小木头 (视频3)
I — Interpretation (自述)
任何完整的 Loop 系统都可以拆解为 5 个组件 + 1 根脊柱:
- 心跳 (Heartbeat): 定时触发器 (cron),到点自动启动循环。
- 工作树 (Work Tree): 隔离的工作空间,多个 agent 并行时不踩踏。
- Skill: 项目规则和约束,写入 SKILL.md,每个 agent 自动读取。
- 连接器 (Connector): 通过 MCP 接入外部工具 (issue 系统、数据库等)。
- 子智能体 (Sub-agent): 拆分生产和审查,避免自产自检。
- 脊柱 (Spine): 持久化记忆层,记录历史操作、尝试结果、待办事项。
脊柱是关键: Agent 会遗忘 (上下文有限),所以必须在对话之外建立持久化记忆。
A1 — Past Application (书中案例)
案例1: Boris 的多 Loop 系统 (视频2)
- 心跳: 每 30 分钟触发 support loop
- Work Tree: 每个 agent 独立分支
- Skill: 项目规则写入 agents.md
- 连接器: MCP 接入 Intercom、Stripe、Supabase
- 子智能体: 写代码和审代码分开
- 脊柱: 共享 signals 文件夹
案例2: 选题 Loop (视频3)
- 心跳: 每天早上 8 点 cron
- Work Tree: 独立分支
- Skill: research + topic-score 技能
- 连接器: research API
- 子智能体: topic-score 作为评级 agent
- 脊柱: inbox.md 文件
A2 — Future Trigger (未来触发)
- 搭建完整 loop 系统时: "我要做一个完整的 X 自动化系统"
- 审计现有系统时: "我的 loop 缺什么组件?"
- 从单 loop 扩展到多 loop 时: "怎么让多个 loop 协同?"
- 团队推广 loop 时: 用这个架构作为"完整 loop 长什么样"的蓝图
语言信号: "设计一个完整的 loop 系统"、"我的 loop 缺什么"、"loop 系统怎么组织"、"多 loop 协同"
与相邻 skill 的区别:
loop-three-elements: 最小可用单元 (本 skill 是完整系统)maker-checker: 子智能体分工的具体实现 (本 skill 是整体架构)loop-build-path: 构建步骤 (本 skill 是静态结构)
E — Execution (可执行步骤)
Step 1: 审计现有系统
对照 5+1 清单,逐项检查:
□ 心跳: 有定时/事件触发吗?
□ Work Tree: 有隔离工作空间吗?
□ Skill: 规则写成了 SKILL.md 吗?
□ 连接器: 接入了外部工具吗?
□ 子智能体: 生产和审查分开了吗?
□ 脊柱: 有持久化记忆吗?
What ships with it
2 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 · 120 lines · 116 tokens per session scan A 1087cba747e6
loop-5plus1-architecture is a skill published in the GitHub repository kangarooking/loop-engineering-skill (23 stars, last pushed 2mo ago), licensed MIT. It adds 116 tokens to every session and 1,550 once invoked, about $0.0006 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.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
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…