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/three-stage-evolutionnpx skills add kangarooking/loop-engineering-skill --skill three-stage-evolutiongit 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 | $0.00107 | $0.01530 |
| Opus 5 | $0.00053 | $0.00765 |
| Sonnet 5 | $0.00021 | $0.00306 |
| Haiku 4.5 | $0.00011 | $0.00153 |
Grade A, and why
three-stage-evolution 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 — 110 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: three-stage-evolution
description: |
评估个人或团队在 AI 工具使用上的进化阶段,给出下一步提升建议。
当用户想判断"我现在 AI 用得怎么样"、"下一步该学什么"、或团队要做 AI 能力评估时使用。
不适用于: 已经处于 Loop 阶段需要优化具体系统的场景。
关键 trigger: "我现在在哪个阶段"、"AI 使用下一步"、"团队 AI 能力评估"。
source_book: "Loop Engineering 视频合集"
source_chapter: 视频3 (小木头) / 视频2 (Boris Cherny)
tags: [evolution, self-assessment, capability-model, growth]
related_skills: [loop-build-path, loop-worthiness-test, comprehension-gap]
三阶段进化模型 — 定位你的 AI 使用阶段
R — Reading (原文)
"第一阶段,逐行驾驶...你手写代码模型给你自动补全。第二个阶段,并行手动,你或许会同时开5个、10个对话,每个都在干活。但每次对话都是你亲手发起的...第三阶段你不再发起对话,你写了一个系统,让他自己去读你的仓库,读issue,读CI的失败,自己决定该用什么样的提示词驱动智能体工作。" — 小木头 (视频3)
I — Interpretation (自述)
个人与 AI 协作方式经历三个进化阶段:
- 逐行驾驶 (Prompting): 手写代码/内容,AI 给自动补全。人是操作者,AI 是工具。
- 并行手动 (Parallel): 同时开 5-10 个对话,每个都在干活。人是调度员,在多个窗口间切换。
- Loop 系统 (Loop Engineering): 不再发起对话,设计一个系统让 AI 自主工作。人是系统设计者。
每个阶段的特征:
- Stage 1: 每次任务都是一次性的,AI 是"高级自动补全"
- Stage 2: 效率提升但人是瓶颈,所有对话都需亲手发起
- Stage 3: 系统自主运行,人只需要设计和监督
A1 — Past Application (书中案例)
案例1: Boris Cherny (视频2)
- 描述了从"逐行驾驶"到"写 loop"的进化
- 当前处于 Stage 3: 有一堆 loop 在跑,自己只写新的 loop
案例2: 小木头 (视频3)
- 自评处于 Stage 2 (并行手动)
- 正在向 Stage 3 过渡 (演示了选题 loop)
案例3: Adam Gillock (视频1)
- 非技术背景,但已经用 loop 做视频剪辑
- 说明 Stage 3 不限于技术人员
A2 — Future Trigger (未来触发)
- 自我评估时: "我现在 AI 用得怎么样?"
- 制定学习计划时: "下一步该学什么?"
- 团队能力建设时: 评估团队整体 AI 使用阶段
- 向他人解释 Loop Engineering 时: 用这个模型说明"为什么要升级到 Stage 3"
语言信号: "我现在在哪个阶段"、"AI 使用下一步"、"团队 AI 能力评估"、"从手动到自动"
与相邻 skill 的区别:
loop-build-path: 升级到 Stage 3 后的构建指南 (本 skill 是定位和决策)loop-worthiness-test: Stage 3 中判断具体任务要不要做 loop (本 skill 是整体阶段评估)comprehension-gap: Stage 3 的风险 (本 skill 是 Stage 3 的进阶路径)
E — Execution (可执行步骤)
Step 1: 自评当前阶段
回答以下问题:
- 你每次用 AI 都是亲手发起对话吗? → Stage 1 或 2
- 你有多个 AI 对话同时运行吗? → Stage 2
- 你有定时/事件触发的 AI 任务吗? → 可能是 Stage 3
- 你设计过"让 AI 自主决定做什么"的系统吗? → Stage 3
Step 2: 对照特征定位
- 如果主要是"人发起→AI 执行→人反馈" → Stage 1
- 如果主要是"人同时管理多个 AI 任务" → Stage 2
- 如果主要是"系统自动运行,人设计系统" → Stage 3
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 · 110 lines · 107 tokens per session scan A d188682ead69
three-stage-evolution is a skill published in the GitHub repository kangarooking/loop-engineering-skill (23 stars, last pushed 2mo ago), licensed MIT. It adds 107 tokens to every session and 1,530 once invoked, about $0.0005 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.
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…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
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…
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…