# Send Writing Style Iterator to your agent
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
## Fast path
- Download the package from Yavira.
- Extract it into a folder your agent can access.
- Paste one of the prompts below and point your agent at the extracted folder.
## Suggested prompts
### New install

```text
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete.
```
### Upgrade existing

```text
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run.
```
## Machine-readable fields
```json
{
  "schemaVersion": "1.0",
  "item": {
    "slug": "writing-style-iterator",
    "name": "Writing Style Iterator",
    "source": "tencent",
    "type": "skill",
    "category": "AI 智能",
    "sourceUrl": "https://clawhub.ai/shanjiaming/writing-style-iterator",
    "canonicalUrl": "https://clawhub.ai/shanjiaming/writing-style-iterator",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadUrl": "/downloads/writing-style-iterator",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=writing-style-iterator",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "packageFormat": "ZIP package",
    "primaryDoc": "SKILL.md",
    "includedAssets": [
      "SKILL.md"
    ],
    "downloadMode": "redirect",
    "sourceHealth": {
      "source": "tencent",
      "slug": "writing-style-iterator",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-05-04T13:28:12.064Z",
      "expiresAt": "2026-05-11T13:28:12.064Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=writing-style-iterator",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=writing-style-iterator",
        "contentDisposition": "attachment; filename=\"writing-style-iterator-0.1.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null,
        "slug": "writing-style-iterator"
      },
      "scope": "item",
      "summary": "Item download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this item.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/writing-style-iterator"
    },
    "validation": {
      "installChecklist": [
        "Use the Yavira download entry.",
        "Review SKILL.md after the package is downloaded.",
        "Confirm the extracted package contains the expected setup assets."
      ],
      "postInstallChecks": [
        "Confirm the extracted package includes the expected docs or setup files.",
        "Validate the skill or prompts are available in your target agent workspace.",
        "Capture any manual follow-up steps the agent could not complete."
      ]
    }
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/writing-style-iterator",
    "downloadUrl": "https://openagent3.xyz/downloads/writing-style-iterator",
    "agentUrl": "https://openagent3.xyz/skills/writing-style-iterator/agent",
    "manifestUrl": "https://openagent3.xyz/skills/writing-style-iterator/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/writing-style-iterator/agent.md"
  }
}
```
## Documentation

### Writing Style Iterator — 写作风格记忆系统

你具备写作风格记忆能力。通过 Writing Style Iterator，你能记住用户的写作风格偏好，并随着每次交互变得更懂用户。

### 数据位置

~/.writing-style-iterator/ 是一个 git 仓库，存放：

style.md — 用户的风格规则文件
drafts/ — 草稿快照（用于 diff 和回滚）

首次使用前，如果目录不存在，先初始化：

mkdir -p ~/.writing-style-iterator/drafts && git -C ~/.writing-style-iterator init && touch ~/.writing-style-iterator/style.md && git -C ~/.writing-style-iterator add . && git -C ~/.writing-style-iterator commit -m "init"

### 首次生成

1. 用户发起写作请求
       ↓
2. 加载style：cat ~/.writing-style-iterator/style.md
       ↓
3. 按 style.md 中的规则生成/修改内容
       ↓
4. 保存草稿（写入用户文件 + 记录快照 + commit）

### 迭代修改（核心循环）

5. 用户在编辑器中修改（你不参与这一步）
       ↓
6. 用户回来找你（可能带反馈，也可能什么都不说）
       ↓
7. 获取 diff，分析用户改了什么
       ↓
8. 提取风格规则 → 写入 style.md → 通知用户
       ↓
9. 判断是否需要更新内容：
       └─ 新规则明显影响其他部分 → 主动修改并保存
       └─ 用户口头要求了修改 → 照做
       └─ 没什么要改的 → 不必产出新内容
       ↓
      回到步骤 5，直到用户满意

### 何时激活

当用户的请求涉及以下任何场景时，加载 style.md 并进入 Writing Style Iterator 工作流：

写文章、博客、评论、文案
改写、润色、调整语气
翻译并调整风格
任何「帮我写/改」类请求

### 操作方法

不需要专门的 CLI 工具。所有操作都是文件操作 + git，用 && 链接保证原子性。

### 加载style

cat ~/.writing-style-iterator/style.md

### 保存草稿

# 将内容写入用户文件，同时记录快照（用绝对路径做目录结构，避免同名冲突）
# 例：/Users/sjm/blog/article.md → ~/.writing-style-iterator/drafts/Users/sjm/blog/article.md
mkdir -p ~/.writing-style-iterator/drafts/$(dirname <用户文件的绝对路径>) && cp <用户文件> ~/.writing-style-iterator/drafts/<用户文件的绝对路径> && git -C ~/.writing-style-iterator add . && git -C ~/.writing-style-iterator commit -m "draft: <文件名>"

每次生成/修改内容都必须做这一步，否则后续 diff 无法工作。

### 获取用户修改

diff ~/.writing-style-iterator/drafts/<用户文件的绝对路径> <用户文件>

输出用户自上次保存以来的所有变更。

### 更新style

# 将新内容写入 style.md 并 commit
cat > ~/.writing-style-iterator/style.md << 'EOF'
(更新后的 style.md 完整内容)
EOF
git -C ~/.writing-style-iterator add style.md && git -C ~/.writing-style-iterator commit -m "style: <修改摘要>"

### 回滚草稿（AI 改坏了用户文件时）

git -C ~/.writing-style-iterator checkout HEAD~1 -- drafts/<用户文件的绝对路径> && cp ~/.writing-style-iterator/drafts/<用户文件的绝对路径> <用户文件>

### 回滚 style.md

git -C ~/.writing-style-iterator checkout HEAD~1 -- style.md

### 查看版本历史

git -C ~/.writing-style-iterator log --oneline

### 核心原则：记录+可能的泛化

记录，不强行泛化 你大概总结出规律了可以泛化一下 style.md

### 不同粒度的修改 → 不同类型的记录

用户做了什么写入  style.md 的什么位置改了一个词/短语忌口清单 → 替换条目删掉了某种句式忌口清单 → 禁用结构重写了一整段参考示例 → before/after 对比加了批注说"太官方了"核心原则 → 补充语气要求

### 输入来源

diff：用户对草稿的修改
行内批注：用户可能在文件里加了标记。没有固定格式——<!-- 太官方了 -->、(这句不好)、甚至就一个 ?，你都应该能识别出来
口头反馈：用户直接和你说的话

### 流程

看 diff + 批注 + 口头反馈
按修改粒度决定写入 style.md 的位置
直接写入 style.md 并 commit
通知用户你做了什么：
已更新style：
。。。
如需撤回，告诉我即可。

不要问 Y/N。 直接做，然后通知。不满意就回滚。

### 内容更新策略

revise 不一定产出更新后的内容。你需要自己判断：

用户改过的部分：保留，不覆盖
用户口头要求了修改：照做
新规则明显影响其他部分：主动应用
没什么要改的：不需要产出新内容，只更新 style.md 就行

核心目标是节省用户能量。能自动做的就自动做。

### 重要原则

节省用户能量，减少交互摩擦：最高原则。规则直接写入不用问，做完通知就行。
忠实记录，不强行泛化：用户的修改是什么就记什么。模式是长期积累自然涌现的。
用户不满意可以撤回：style.md 和草稿都有完整版本历史，随时回滚。
diff 是事实：diff 告诉你用户实际做了什么，比你的猜测更可靠。
用户的意图高于一切：用户怎么说就怎么做，Writing Style Iterator 的工作流是辅助而非限制。
## Trust
- Source: tencent
- Verification: Indexed source record
- Publisher: shanjiaming
- Version: 0.1.0
## Source health
- Status: healthy
- Item download looks usable.
- Yavira can redirect you to the upstream package for this item.
- Health scope: item
- Reason: direct_download_ok
- Checked at: 2026-05-04T13:28:12.064Z
- Expires at: 2026-05-11T13:28:12.064Z
- Recommended action: Download for OpenClaw
## Links
- [Detail page](https://openagent3.xyz/skills/writing-style-iterator)
- [Send to Agent page](https://openagent3.xyz/skills/writing-style-iterator/agent)
- [JSON manifest](https://openagent3.xyz/skills/writing-style-iterator/agent.json)
- [Markdown brief](https://openagent3.xyz/skills/writing-style-iterator/agent.md)
- [Download page](https://openagent3.xyz/downloads/writing-style-iterator)