conejo-code

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Conejo-code — active coding

Conejo-code — 主动编码

The hands-on coding loop. Philosophy lives in [[conejo]]; this skill is how you execute it. Relentless, disciplined, experiment-driven and detail-oriented coding practice. ALWAYS set a timer to (1) wait for the user, (2) wait for Claude agents, (3) Claude or any other tool or (4) wait for opencode or forge agents. NEVER expect that you will be awaken by the response of any of these. YOU MUST BE FULLY AUTONOMOUS AND IMPLEMENT THE ORIGINAL PLAN INFERRING THE BEST ALTERNATIVES TO ENSURE IT WORKS.

实操编码循环。理念源自[[conejo]]；本技能是其落地执行方式。这是一种严谨、自律、以实验为导向且注重细节的编码实践。务必设置计时器，用于：(1) 等待用户响应，(2) 等待Claude Agent响应，(3) 等待Claude或其他工具响应，(4) 等待opencode或forge Agent响应。切勿依赖上述对象的响应来唤醒你。你必须完全自主执行原始规划，推断最佳替代方案以确保任务完成。

The loop (adapted feature stages)

循环流程（适配后的功能开发阶段）

Brainstorm (Specs) → run "opencode run $PROMPT --model zai-coding-plan/glm-5.1 --dangerously-skip-permissions --dir path/to/this/dir" to review your specs → Plan "opencode run $PROMPT --model deepseek/deepseek-v4-pro --dangerously-skip-permissions --dir path/to/this/dir" to review your plan (glm-5.1 is smarter but slow, deepseek is a bit faster) → Interface → "opencode run $PROMPT --model deepseek/deepseek-v4-pro --dangerously-skip-permissions --dir path/to/a/clone/of/this/dir" to create tests failing tests → "opencode run $PROMPT --model deepseek/deepseek-v4-pro --dangerously-skip-permissions --dir path/to/a/clone/of/this/dir" to Implement → Improve tests → Code review → (loop).
Full stage notes:
```
references/feature-stages.md
```
. UI stages (design/UIUX-review) and any visual work are owned by [[conejo-frontend]].
Attention! Don't be afraid of sending several agents: split the tasks as much as possible.
PR over PR ALWAYS. Push. After it is at Github, clean their mess.
PRESERVE your context no matter what! Send these other agents first, so we can have different perspectives, but if you need to do the work, send your agents. Don't pollute your context.

头脑风暴（需求规格）→ 运行 "opencode run $PROMPT --model zai-coding-plan/glm-5.1 --dangerously-skip-permissions --dir path/to/this/dir" 评审需求规格 → 规划阶段运行 "opencode run $PROMPT --model deepseek/deepseek-v4-pro --dangerously-skip-permissions --dir path/to/this/dir" 评审规划（glm-5.1更智能但速度慢，deepseek速度稍快）→ 接口设计 → 运行 "opencode run $PROMPT --model deepseek/deepseek-v4-pro --dangerously-skip-permissions --dir path/to/a/clone/of/this/dir" 创建失败测试用例 → 运行 "opencode run $PROMPT --model deepseek/deepseek-v4-pro --dangerously-skip-permissions --dir path/to/a/clone/of/this/dir" 实现功能 → 优化测试用例 → 代码评审 →（循环）。
完整阶段说明：
```
references/feature-stages.md
```
。UI阶段（设计/UIUX评审）及所有可视化工作由[[conejo-frontend]]负责。
注意！不要害怕调用多个Agent：尽可能拆分任务。
始终采用PR叠加PR的方式。推送代码。代码上传至Github后，清理冗余内容。
无论如何都要保留上下文！先调用其他Agent获取不同视角，但如果需要执行具体工作，调用你的专属Agent。不要污染自身上下文。

Testing — READ FIRST

测试 — 请先阅读

The universal red-green doctrine is canonical in

references/testing-doctrine.md

. Backend/logic/CLI/library use

vitest

; UI behavior uses agent-browser click-flows (see [[conejo-frontend]]). Never implement before a failing test.

通用红绿准则的权威说明见

references/testing-doctrine.md

。后端/逻辑/CLI/库使用

vitest

；UI行为测试使用agent-browser点击流（详见[[conejo-frontend]]）。绝不在编写失败测试用例前实现功能。

Standards & helpers

标准与辅助工具

Coding standards:

references/coding-standards.md

-api-design.md

Dependency hygiene (bun):
```
references/deps-check.md
```
.
Capture bug-fix learnings:
```
references/bug-fix-learning.md
```
.
Strengthen tests after green:
```
references/auto-improve-tests.md
```
.
Optional auto-trigger hooks:
```
hooks/
```
— install with
```
scripts/install-conejo-hooks.sh
```
.

编码标准：

references/coding-standards.md

（含

-api-design.md

）。

依赖管理（bun）：
```
references/deps-check.md
```
。
Bug修复学习记录：
```
references/bug-fix-learning.md
```
。
测试用例优化：
```
references/auto-improve-tests.md
```
。
可选自动触发钩子：
```
hooks/
```
— 通过
```
scripts/install-conejo-hooks.sh
```
安装。

Tooling

工具栈

bun (not npm), vitest (not jest, not

bun test

), vite-plus (

vp

) for build/check. ESM only.

使用bun（而非npm）、vitest（而非jest或

bun test

）、vite-plus（

vp

）进行构建/检查。仅支持ESM。