← 返回 Skills 市场
xjouska

AI-CMO Social Media Promoter

作者 Xiaosen Li · GitHub ↗ · v1.0.3 · MIT-0
cross-platform ✓ 安全检测通过
136
总下载
0
收藏
0
当前安装
4
版本数
在 OpenClaw 中安装
/install ai-cmo
功能描述
Find the best social media promotion spots for any product and generate platform-tailored engagement copy
使用说明 (SKILL.md)

AI-CMO: Social Media Promotion Intelligence

AI-CMO analyzes any product URL or name and finds the best promotion opportunities across Reddit, YouTube, Twitter/X, Quora, Product Hunt, and Hacker News. It returns ranked posts with AI-generated engagement copy tailored to each platform's tone.

This skill calls the AI-CMO REST API to retrieve promotion data. It does not require a local browser — all analysis happens server-side. The agent uses the returned data to help you engage on social media.

Homepage: \x3Chttps://www.aicmo.site> API Docs: \x3Chttps://www.aicmo.site/api-docs>

Runtime requirements

  • OS: macOS · Linux · Windows
  • Env: AICMO_API_KEY — your AI-CMO API key (get one at \x3Chttps://www.aicmo.site> → Settings → API Keys)

When to use

  • You want to find high-value posts/threads to engage with for a product
  • You need AI-generated comments customized per platform
  • You want a CSV of promotion targets for batch processing

When not to use

  • You only need a simple Google search (use a search engine instead)
  • The product has no web presence yet (AI-CMO needs existing content to find)

Prerequisites

  • An AI-CMO account — register at \x3Chttps://www.aicmo.site>
  • Environment variable: AICMO_API_KEY

Get an API key

  1. Log in to \x3Chttps://aicmo.site>
  2. Go to Settings → API Keys
  3. Click Create Key and save the key securely
export AICMO_API_KEY="sk-cmo-your-key-here"

Quick start: analyze a product

Analyze a product and get ranked promotion spots. Consumes 1 credit per call.

Note: Analysis takes 2–4 minutes (scraping 6 platforms + AI scoring). Set your HTTP client timeout to at least 300 seconds.

const resp = await fetch("https://www.aicmo.site/api/v1/open/analyze", {
  method: "POST",
  headers: {
    "Content-Type": "application/json",
    "x-api-key": process.env.AICMO_API_KEY!,
  },
  body: JSON.stringify({
    product: "https://your-product.com",
    language: "en",
    max_results: 10,
  }),
});

const result = await resp.json();

for (const spot of result.promotions) {
  console.log(`#${spot.rank} [${spot.platform}] ${spot.title}`);
  console.log(`  URL: ${spot.url}`);
  console.log(`  Comment: ${spot.suggested_comment}`);
  console.log(`  Score: ${spot.composite_score} | Risk: ${spot.risk_level}`);
}

The response contains promotions[] (ranked posts with url, suggested_comment, platform, composite_score, risk_level, etc.) and strategy (action plan). For CSV output, use /analyze/csv instead — same request, returns a downloadable CSV file.

Base URL: https://www.aicmo.site/api/v1/open (always use www.aicmo.site).

CSV columns

Column Description
Rank Priority ranking (1 = best)
Platform reddit, youtube, twitter, quora, producthunt, hackernews
URL Direct link to the post/thread
Title Post title
Suggested Comment AI-generated engagement copy for this specific post
Comment Strategy Context on how to approach the conversation
Why Recommend Why this post is a good promotion opportunity
Best Timing Recommended engagement window
Risk Level low, medium, or high
Estimated Exposure Approximate reach
Score Composite quality score (0–100)

Engaging with results

After receiving promotion spots, the agent can help you engage by:

  1. Browsing to the post URL
  2. Reading the existing discussion for context
  3. Composing a reply using the suggested_comment as a starting point
  4. Adapting the tone to match the thread's conversation style

Important: Always review the suggested comment before posting. The AI provides a strong starting point, but human judgment ensures the reply feels authentic and adds genuine value to the conversation.

References (deep dive)


中文版 / Chinese Translation


AI-CMO:社媒推广智能分析

AI-CMO 可分析任意产品 URL 或名称,在 Reddit、YouTube、Twitter/X、Quora、Product Hunt 和 Hacker News 上找到最佳推广机会。返回排名后的帖子列表,并为每个帖子生成匹配该平台风格的 AI 互动文案。

本 Skill 调用 AI-CMO REST API 获取推广数据,不需要本地浏览器——所有分析在服务端完成。

何时使用

  • 为产品找到高价值的帖子/讨论串进行互动
  • 获取为每个平台定制的 AI 评论文案
  • 获取 CSV 格式的推广目标清单用于批量处理

何时不使用

  • 只需要简单搜索(请直接用搜索引擎)
  • 产品还没有任何网络信息(AI-CMO 需要已有内容才能分析)

前置条件

  • AI-CMO 账号——在 \x3Chttps://www.aicmo.site> 注册
  • 环境变量:AICMO_API_KEY(在 设置 → API 密钥 中创建)

代码示例与上方英文部分一致,此处不再重复。

核心接口

接口 说明 消耗积分
GET /quota 查询剩余积分
POST /analyze 分析产品,返回 JSON 推广位 1 积分
POST /analyze/csv 分析产品,返回 CSV 文件 1 积分
GET /history 分析历史列表
GET /history/{id}/csv 导出过往分析为 CSV

使用结果进行互动

Agent 可以帮你:浏览帖子 URL → 阅读上下文 → 以 suggested_comment 为起点撰写回复 → 调整语气匹配对话风格。

发帖前务必审阅。AI 提供的是起点,人工判断确保回复真实自然。

参考文档

安全使用建议
This skill appears internally consistent: it calls a remote AI-CMO service and needs only an API key. Before installing, verify the provider (https://www.aicmo.site) is legitimate and that the API key is stored securely. Understand that the agent may be guided to browse and read public posts to help compose replies — review any AI-generated comments before posting to avoid spam, policy violations, or reputational harm. Also confirm pricing/credit costs and that the returned recommendations comply with the terms of service of the target platforms. The only minor metadata gap: the registry entry shows no homepage while SKILL.md references https://www.aicmo.site; you may want to confirm the publisher and service identity prior to use.
能力评估
Purpose & Capability
The name/description map to calling a remote promotion-analysis service. The only declared secret (AICMO_API_KEY) matches that purpose. No unrelated binaries, config paths, or credentials are requested.
Instruction Scope
SKILL.md consistently directs the agent to call the external API (https://www.aicmo.site) and use returned promotion spots. It also suggests the agent may 'browse to the post URL' and 'read the discussion' to compose replies — this is reasonable for crafting comments, but is broader than the pure analysis step and means the agent may access web content during follow‑up actions. Also, the registry metadata earlier listed no homepage yet SKILL.md includes https://www.aicmo.site (minor metadata inconsistency).
Install Mechanism
Instruction-only skill with no install spec and no code files; nothing is written to disk and no external packages are pulled.
Credentials
Only one env var (AICMO_API_KEY) is required, which is proportionate to calling the stated API. No other secrets or config paths are requested.
Persistence & Privilege
Defaults are used (not always:true). The skill does not request permanent or elevated platform privileges. It describes external API usage and optional browsing, but does not assert self-enabling or system-wide changes.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install ai-cmo
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /ai-cmo 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.3
- Updated all site and API URLs to use https://www.aicmo.site (with www). - Added a note about analysis time: processing takes 2–4 minutes and HTTP client timeout should be set to at least 300 seconds. - Clarified API base URL: always use `https://www.aicmo.site/api/v1/open`. - No code or feature changes; documentation updates only.
v1.0.2
No functional changes were made in this version. - Added a standardized skill header with name, description, and required environment variable. - No other documentation or code updates detected.
v1.0.1
- Added explicit links to the AI-CMO homepage and API documentation for easier access. - Specified supported operating systems and clarified API key environment variable setup in a new "Runtime requirements" section. - Updated reference links to use markdown formatting. - No feature changes; documentation improvements only.
v1.0.0
Initial release of AI-CMO social media promotion skill. - Analyze any product URL or name to find top promotion opportunities on Reddit, YouTube, Twitter/X, Quora, Product Hunt, and Hacker News. - Returns ranked posts with AI-generated, platform-tailored engagement copy. - Supports result export as CSV for batch processing. - Detailed API reference, platform tips, and scheduling guidance included. - Chinese-language documentation provided.
元数据
Slug ai-cmo
版本 1.0.3
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 4
常见问题

AI-CMO Social Media Promoter 是什么?

Find the best social media promotion spots for any product and generate platform-tailored engagement copy. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 136 次。

如何安装 AI-CMO Social Media Promoter?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install ai-cmo」即可一键安装,无需额外配置。

AI-CMO Social Media Promoter 是免费的吗?

是的,AI-CMO Social Media Promoter 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

AI-CMO Social Media Promoter 支持哪些平台?

AI-CMO Social Media Promoter 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 AI-CMO Social Media Promoter?

由 Xiaosen Li(@xjouska)开发并维护,当前版本 v1.0.3。

💬 留言讨论