Back to Details

plan

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

/plan

/plan

Write a plan file for the current task. Requires /orient to have run first.
为当前任务编写计划文件。需先执行/orient步骤。

Sequence

流程

  1. Check prerequisites. Confirm a worktree exists and /orient has identified the task. If not, stop and tell the user.
  2. Brainstorm (one round). Invoke 
    /brainstorming
    with the task description. One round only — capture direction, do not iterate.
  3. Read the template. Read 
    ai-workspace/plans/TEMPLATE.md
    from the repo root.
  4. Write the plan. Create 
    ai-workspace/plans/<branch-slug>.md
    with ALL template fields:
FieldHow to fill
BranchCurrent branch name
CreatedToday's date (YYYY-MM-DD)
Status
In Progress
Threat modelSee selection table below
Scope ceilingKeep template defaults (400/6 soft, 800/10 hard)
Task1-3 sentences: what and why
StepsCheckbox list of concrete implementation steps
Confidence ScaffoldRequired for 
adversarial
. Recommended for complex 
advisory
.
Outcomes & LearningsLeave empty — populated by /archive
  1. 检查前置条件：确认工作树已存在且/orient已识别任务。若未满足，停止操作并告知用户。
  2. 头脑风暴（一轮）：调用/brainstorming工具并传入任务描述。仅进行一轮——确定方向即可，无需迭代。
  3. 读取模板：从仓库根目录读取
    ai-workspace/plans/TEMPLATE.md
    文件。
  4. 编写计划：创建
    ai-workspace/plans/<branch-slug>.md
    文件，需填充所有模板字段：
字段填充方式
Branch当前分支名称
Created今日日期（YYYY-MM-DD格式）
Status
In Progress
Threat model参考下方选择表
Scope ceiling保留模板默认值（软限制400/6，硬限制800/10）
Task1-3句话：说明任务内容及目的
Steps具体实施步骤的复选框列表
Confidence Scaffold
adversarial
类型任务必填，复杂
advisory
类型任务推荐填写
Outcomes & Learnings留空——由/archive步骤填充

Threat model selection

威胁模型选择

SignalModel
Internal tooling, refactor, docs, tests, config
advisory
Auth, secrets, input validation, CI, hooks, permissions
adversarial
Unsure
adversarial
信号模型
内部工具、重构、文档、测试、配置
advisory
认证、密钥、输入验证、CI、钩子、权限
adversarial
不确定
adversarial

Guardrails

约束规则

  • Do NOT review the plan. That is /review (Step 4).
  • Do NOT start implementation. The plan is the deliverable.
  • Do NOT skip any template field.
  • If brainstorming reveals the task is one-sentence scope, say so and skip.
  • Scope ceiling values are fixed — do not change them.
  • Steps must be concrete actions with checkboxes, not vague phases.
  • Include test steps explicitly.
  • 不得评审计划——评审为/review（第4步）的内容。
  • 不得开始实施——计划即为交付成果。
  • 不得跳过任何模板字段。
  • 若头脑风暴发现任务范围仅为一句话，需告知用户并跳过此步骤。
  • 范围上限值固定——不得修改。
  • 步骤必须是带复选框的具体操作，而非模糊阶段。
  • 需明确包含测试步骤。

Output

输出

After writing, report:
Plan written: ai-workspace/plans/<name>.md
Threat model: advisory|adversarial
Steps: <count>
Next step: /review
编写完成后，输出如下内容：
Plan written: ai-workspace/plans/<name>.md
Threat model: advisory|adversarial
Steps: <count>
Next step: /review