社交媒体智能回复助手

v1.0.0

社交媒体智能回复助手。为内容创作者批量生成高质量评论回复,提升账号互动率。 支持多平台风格适配(小红书、抖音、微博、B站、知乎),多种回复策略(幽默、专业、温暖、高情商、引导互动)。 可维护个人风格模板,一键生成符合人设的回复。 当用户说"帮我回评论"、"回复评论"、"生成回复"、"评论回复"、"怎么回这条评论"...

0· 72·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for freedompixels/cn-social-reply.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "社交媒体智能回复助手" (freedompixels/cn-social-reply) from ClawHub.
Skill page: https://clawhub.ai/freedompixels/cn-social-reply
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install cn-social-reply

ClawHub CLI

Package manager switcher

npx clawhub@latest install cn-social-reply
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
The name/description state it generates social-media replies; the SKILL.md and included script implement exactly that (generate replies via the agent and manage local style templates). No unrelated credentials, cloud APIs, or unrelated binaries are requested.
Instruction Scope
Instructions focus on analyzing comments, selecting platform/style, generating replies, and managing local templates. They do not instruct reading other user files, system state, shell history, or sending data to external endpoints. The only file access is the styles.json template file in the user's home workspace.
Install Mechanism
There is no install spec (instruction-only plus a small helper script). The included Python script is small, readable, and only performs local JSON read/write. No downloads, package installs, or archive extraction are present.
Credentials
The skill declares no required environment variables or credentials, which is appropriate. Minor note: SKILL.md invokes 'python3' to run scripts but the manifest did not list a required binary; this is a small operational mismatch (it assumes a Python 3 runtime is available).
Persistence & Privilege
The skill does not request always:true or any special persistent privileges. It stores templates only under ~/.qclaw/workspace/cn-social-reply/styles.json, which is a scoped, user-local path and not a system-wide configuration.
Assessment
This skill appears to do what it says: generate reply text and let you save/load personal style templates locally. Before installing, consider: 1) Ensure you have Python 3 available (SKILL.md calls python3). 2) Templates and any text you save are stored locally at ~/.qclaw/workspace/cn-social-reply/styles.json — review that file if it will contain sensitive information. 3) The agent will see comment text when invoked, so avoid sending PII to the skill or enabling it to run autonomously on unvetted inputs. 4) Because the skill can be invoked by trigger phrases, confirm you are comfortable with the agent having permission to call it; there are no hidden network calls in the provided code. If you need stronger guarantees, run the style_manager.py locally yourself to inspect behavior and file location before enabling the skill.

Like a lobster shell, security has layers — review code before you run it.

Runtime requirements

💬 Clawdis
latestvk9762rzwgpbgetx1mvkrgnx6e984xyf0
72downloads
0stars
1versions
Updated 1w ago
v1.0.0
MIT-0

社交媒体智能回复助手

为内容创作者智能生成评论回复,提升互动率、维护粉丝关系。

🎯 核心能力

功能说明
单条回复输入评论 + 上下文,生成精准回复
批量回复一次处理多条评论,统一风格
风格切换幽默 / 专业 / 温暖 / 高情商 / 互动引导 / 毒舌俏皮
平台适配自动匹配平台调性(小红书emoji风 / 抖音口语化 / 知乎理性风等)
风格模板保存/加载个人回复风格偏好

📋 使用方式

方式一:直接给出评论

用户:帮我回这条评论 "你皮肤真好,用的什么护肤品?"

Agent 会分析评论意图,直接生成合适回复。

方式二:指定平台+风格

用户:小红书评论回复,幽默风格:
1. "姐妹你这个妆也太好看了吧!"
2. "求教程!急!"
3. "口红什么色号啊"

Agent 会按指定平台风格批量生成回复。

方式三:附带帖子上下文

用户:我刚发了一篇测评iPhone 16的文章,这几条评论帮我回一下:
1. "电池续航真的提升了吗?"
2. "对比安卓优势在哪?"
3. "值不值得从14升级?"

方式四:管理风格模板

用户:保存我的回复风格 - 偏专业但不枯燥,偶尔抛梗,字数控制在20字以内
用户:加载我的回复风格

🧠 回复策略详解

1. 幽默风格 😄

  • 适度玩梗,自嘲/调侃,拉近距离
  • 例:评论"太厉害了吧" → "主要是脸皮够厚😂 试试你也行"

2. 专业风格 🧐

  • 认真回答,展现专业度,适当科普
  • 例:评论"这个真的有用吗" → "从成分角度看,XX确实有XX临床数据支撑,但效果因人而异"

3. 温暖风格 🥰

  • 走心回应,表达感谢,建立情感连接
  • 例:评论"好喜欢你的内容" → "这句话是我今天收到最好的礼物,谢谢你在❤️"

4. 高情商风格 🎯

  • 巧妙化解尴尬,照顾多方感受,格局打开
  • 例:评论"我觉得你这个不如XX" → "XX确实很棒!我也要多学习,各有各的好✨"

5. 互动引导风格 🚀

  • 回复同时引导更多互动(提问、投票、@好友)
  • 例:评论"这也太好看了" → "你更喜欢A还是B?评论区告诉我~"

6. 毒舌俏皮风格 😏

  • 带点小刺但有趣,适合个人风格鲜明的博主
  • 例:评论"你怎么什么都会" → "会的多是因为……选择困难症😂"

📱 平台风格适配

平台调性特征emoji使用字数建议
小红书闺蜜感、热情、emoji丰富多,姐妹/宝/亲15-40字
抖音口语化、接地气、网感中等,哈哈/笑死10-30字
微博简洁犀利、时事感少,偶尔一个10-25字
B站二次元、梗多、真诚中等,awsl/草15-35字
知乎理性、有逻辑、深度极少30-80字

🔄 工作流程

Agent 收到回复请求后,按以下步骤执行:

  1. 分析评论:识别评论类型(提问/夸赞/吐槽/质疑/无关)
  2. 确定平台:用户指定或从上下文推断,默认小红书
  3. 选择风格:用户指定或从已保存模板加载,默认温暖风格
  4. 生成回复:按平台+风格生成 1-3 个候选回复
  5. 展示结果:带编号列出,用户可要求调整

📁 数据存储

使用 scripts/style_manager.py 管理风格模板:

# 保存风格
python3 {SKILL_DIR}/scripts/style_manager.py save --name "我的风格" --style "幽默" --extra "字数20字以内,偶尔抛梗,喜欢用[doge]表情"

# 列出已保存的风格
python3 {SKILL_DIR}/scripts/style_manager.py list

# 加载风格
python3 {SKILL_DIR}/scripts/style_manager.py load --name "我的风格"

# 删除风格
python3 {SKILL_DIR}/scripts/style_manager.py delete --name "我的风格"

风格数据存储在 ~/.qclaw/workspace/cn-social-reply/styles.json

⚡ 快速开始示例

最简用法:

帮我回这条评论:"写得真好!关注了!"

进阶用法:

小红书风格,帮我批量回复:
1. "姐妹你也太会穿了吧"
2. "链接链接链接!"
3. "这是P过的吧"
4. "请问多高呀"

人设锁定:

我是一个科技博主,风格偏专业但轻松,帮我回:
"这篇文章不太客观吧,感觉在给苹果洗地"

⚠️ 注意事项

  • 回复应该像真人,避免机器人感(不要每次都"谢谢您的关注")
  • 遇到恶意评论/杠精,建议高情商化解而非对骂
  • 不确定回复是否合适时,给用户多个选项让其选择
  • 尊重用户人设,不要生成与博主风格冲突的回复

Comments

Loading comments...