Skip to content
Agentmart
Submit a tool

sa-drafting

当在当前会话中执行具有独立任务的撰写计划时使用

Skill by kun4ai

IndexedMIT★ 7
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/kun4ai-sa-drafting

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

当在当前会话中执行具有独立任务的撰写计划时使用

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/sa-drafting

What's inside

Skill name
sa-drafting
Description
当在当前会话中执行具有独立任务的撰写计划时使用
License
MIT
Instructions
149 lines in SKILL.md · view on GitHub

Files (4)

  • SKILL.md
  • argument-checker-prompt.md
  • drafter-prompt.md
  • style-reviewer-prompt.md

SKILL.md (first lines)

# 子智能体驱动写作

通过为每个任务分派新的子智能体来执行计划,并在每个任务后进行两阶段审查:首先是论证合规性审查,然后是行文质量审查。

**核心原则:** 每个任务使用新鲜的子智能体 + 两阶段审查(先论证后行文)= 高质量、快速迭代。

## 何时使用

```dot
digraph when_to_use {
    "有撰写计划吗?" [shape=diamond];
    "任务大多独立吗?" [shape=diamond];
    "留在此会话中?" [shape=diamond];
    "sa-drafting" [shape=box];
    "sa-execution" [shape=box];
    "先手动执行或头脑风暴" [shape=box];

    "有撰写计划吗?" -> "任务大多独立吗?" [label="是"];
    "有撰写计划吗?" -> "先手动执行或头脑风暴" [label="否"];
    "任务大多独立吗?" -> "留在此会话中?" [label="是"];
    "任务大多独立吗?" -> "先手动执行或头脑风暴" [label="否 - 紧密耦合"];
    "留在此会话中?" -> "sa-drafting" [label="是"];
    "留在此会话中?" -> "sa-execution" [label="否 - 并行会话"];
}
```

**对比 执行计划 (并行会话):**
- 同一会话(无上下文切换)
- 每个任务使用新鲜的子智能体(无上下文污染)
- 每个任务后的两阶段审查:先论证合规,再行文质量
- 更快的迭代(任务之间无需人工介入)

## 流程

```dot
digraph process {
    rankdir=TB;

    subgraph cluster_per_task {
        label="每个任务";
        "分派撰写子智能体 (./drafter-prompt.md)" [shape=box];
        "撰写子智能体提问?" [shape=diamond];
        "回答问题,提供背景" [shape=box];
        "撰写子智能体撰写、自查、提交" [shape=box];
        "分派论证审查子智能体 (./argument-checker-prompt.md)" [shape=box];
        "论证审查子智能体确认草稿符合大纲?" [shape=diamond];
        "撰写子智能体修复论证差距" [shape=box];
        "分派行文质量审查子智能体 (./style-reviewer-prompt.md)" [shape=box];
        "行文质量审查子智能体批准?" [shape=diamond];
        "撰写子智能体修复质量问题" [shape=box];
        "在 TodoWrite 中标记任务完成" [shape=box];
    }

    "阅读计划,提取所有任务全文,记录背景,创建 TodoWrite" [shape=box];
    "还有任务吗?" [shape=diamond];
    "分派最终全文审查子智能体" [shape=box];
    "使用 superpowers:sa-finalization" [shape=box style=filled fillcolor=lightgreen];

    "阅读计划,提取所有任务全文,记录背景,创建 TodoWrite" -> "分派撰写子智能体 (./drafter-prompt.md)";
    "分派撰写子智能体 (./drafter-prompt.md)" -> "撰写子智能体提问?";
    "撰写子智能体提问?" -> "回答问题,提供背景" [label="是"];
    "回答问题,提供背景" -> "分派撰写子智能体 (./drafter-prompt.md)";
    "撰写子智能体提问?" -> "撰写子智能体撰写、自查、提交" [label="否"];
    "撰写子智能体撰写、自查、提交" -> "分派论证审查子智能体 (./argument-checker-prompt.md)";
    "分派论证审查子智能体 (./argument-checker-prompt.md)" -> "论证审查子智能体确认草稿符合大纲?";
    "论证审查子智能体确认草稿符合大纲?" -> "撰写子智能体修复论证差距" [label="否"];
    "撰写子智能体修复论证差距" -> "分派论证审查子智能体 (./argument-checker-prompt.md)" [label="重审"];
    "论证审查子智能体确认草稿符合大纲?" -> "分派行文质量审查子智能体 (./style-reviewer-prompt.md)" [label="是"];
    "分派行文质量审查子智能体 (./style-reviewer-prompt.md)" -> "行文质量审查子智能体批准?";
    "行文质量审查子智能体批准?" -> "撰写子智能体修复质量问题" [label="否"];
    "撰写子智能体修复质量问题" -> "分派行文质量审查子智能体 (./style-reviewer-prompt.md)" [label="重审"];
    "行文质量审查子智能体批准?" -> "在 TodoWrite 中标记任务完成" [label="是"];
    "在 TodoWrite 中标记任务完成" -> "还有任务吗?";
    "还有任务吗?" -> "分派撰写子智能体 (./drafter-prompt.md)" [label="是"];
    "还有任务吗?" -> "分派最终全文审查子智能体" [label="否"];
    "分派最终全文审查子智能体" -> "使用 superpowers:sa-finalization";
}
```

## 提示词模板

- `./drafter-prompt.md` - 分派撰写子智能体
- `./argument-checker-prompt.md` - 分派论证合规审查子智能体
- `./style-reviewer-prompt.md` - 分派行文质量审查子智能体

## 优势

**对比 手动执行:**
- 子智能体自然遵循 ADW (论证驱动写作)
- 每个任务新鲜背景(无混淆)
- 并行安全(子智能体不干扰)
- 子智能体可以提问(工作前和工作中)

**对比 执行计划:**
- 同一会话(无移交)
- 持续进展(无等待)
- 自动审查检查点

**质量门控:**
- 移交前自查捕捉问题
- 两阶段审查:先论证合规,再行文质量
- 审查循环确保修复有效
- 论证合规防止过度/不足论证
- 行文质量确保草稿结构良好

## 危险信号

**绝不:**
- 跳过审查(论证合规 或 行文质量)
- 在问题未修复时继续
- 并行分派多个撰写子智能体(冲突)
- 让子智能体阅读计划文件(应提供全文)
- 跳过场景设置背景(子智能体需要了解任务位置)
- 忽略子智能体问题(先回答再让他们继续)
- 接受“差不多”的论证合规(审查员发现问题 = 未完成)
- 跳过审查循环(审查员发现问题 = 撰写者修复 = 再次审查)
- 让撰写者自查取代实际审查(两者都需要)
- **在论证合规 ✅ 之前开始行文质量审查**(顺序错误)
- 在任一审查有未决问题时移动到下一个任务

From kun4ai/superpowers-academic, MIT.

More from kun4ai/superpowers-academic

Signals

Stars
7
Forks
1
Open issues
0
Last commit (this folder)
Jan 23, 2026 (9 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from kun4ai

Cite this page

sa-drafting skill. Agent Mart. https://agentmart.dev/skills/kun4ai-sa-drafting (last updated 2026-10-24)