跳到主内容
智客 ZICQ

技能库 智客分类:媒体内容 marketing-pipeline-ai-content-automation

Marketing Pipeline Ai Content Automation

使用 AI( Claude/ OpenAI) 和 Remotion 自动将端到端内容创建从研究到视频生成

724 安装量

官方网址:skills.sh

技能介绍

先看中文介绍;官方 description 原文单独保留,不改写 SKILL.md。

做什么

使用 AI( Claude/ OpenAI) 和 Remotion 自动将端到端内容创建从研究到视频生成

何时用

官方 description 未单独写出 Use when。按规范,代理会在用户任务与这段 description 的关键词匹配时激活本技能。

代理如何加载

按 Agent Skills 渐进披露:启动时只加载 name 与 description(约 100 token);任务匹配后才读入整份 SKILL.md 正文;scripts/、references/、assets/ 仅在需要时再读。 本文件正文结构:Marketing Pipeline AI Content Automation、Overview、Installation、Prerequisites、Setup、Clone the repository。

文件分析

文件分析:这是一份仅含 SKILL.md 的指令型技能,代理激活后整份正文进入上下文。 正文约 549 行,超过规范建议的 500 行,激活时仍会整份载入。

官方 description(原文)

Automate end-to-end content creation from research to video generation using AI (Claude/OpenAI) and Remotion

Marketing Pipeline AI Content AutomationOverviewInstallationPrerequisitesSetupClone the repositoryInstall dependenciesCopy environment templateEnvironment ConfigurationAI ServicesResearch APIsDatabase (if applicable)

来源分类:skills.sh agent-skill

SKILL.md 与 Agent 调用

官方规范 ↗
name
marketing-pipeline-ai-content-automation
description
Automate end-to-end content creation from research to video generation using AI (Claude/OpenAI) and Remotion
  1. 发现技能客户端向 Agent 提供名称与描述目录。
  2. 匹配与调用用户指定或任务匹配后,载入 SKILL.md 指令。
  3. 按需加载按步骤读取参考文档、使用脚本与素材。

具体调用语法与可用工具以目标 Agent 客户端为准。 查看调用机制说明 ↗

安装这个技能

Skills CLI ↗

先选择目标 Agent 和安装范围,保留技能包的附属文件,安装后检查客户端能否发现该技能。

交给 Agent 安装

复制安装指令给支持 Agent Skills 的代理,确认其中的目标目录与客户端匹配。

把 Agent Skill「marketing-pipeline-ai-content-automation」安装到我的项目:SKILL.md 原文与官方 description 见 https://zicq.com/zh/skills/skl-12279a7e21e02b30-%E8%90%A5%E9%94%80%E7%AE%A1%E9%81%93-Ai-%E5%86%85%E5%AE%B9%E8%87%AA%E5%8A%A8%E5%8C%96.html
请存为 .cursor/skills/marketing-pipeline-ai-content-automation/SKILL.md 或 .claude/skills/marketing-pipeline-ai-content-automation/SKILL.md,frontmatter 的 name 与 description 保持原样,不要改写。

GitHub 完整包 ↗

终端安装 · Skills CLI

需要 Node.js 与 npx。先查看仓库技能列表,确认实际名称。

npx skills add 'https://github.com/reason-machines/marketing-skills' --list

npx skills add 'https://github.com/reason-machines/marketing-skills' --skill 'marketing-pipeline-ai-content-automation'

CLI 会交互选择目标 Agent,默认安装到项目;用户级安装使用 -g。先通过查看命令核对仓库内容,再用 npx skills list 检查已安装技能。

阅读排版
--- name: marketing-pipeline-ai-content-automation description: Automate end-to-end content creation from research to video generation using AI (Claude/OpenAI) and Remotion triggers: - how do I automate content creation with AI - set up an AI content pipeline for marketing - generate blog posts and videos automatically - use Claude and OpenAI for content automation - create automated content workflow with research - build AI-powered marketing content system - automate video generation from written content - set up content automation with Remotion --- # Marketing Pipeline AI Content Automation > Skill by [ara.so](https://ara.so) — Marketing Skills collection. ## Overview Marketing Pipeline is a comprehensive TypeScript-based content automation system that creates a complete content production pipeline. It automatically: 1. **Research** - Scrapes and analyzes real-time data from sources like TechCrunch, a16z, Twitter, and LinkedIn 2. **Content Generation** - Uses Claude 3 and OpenAI to create content in multiple formats (Top lists, POV, Case Studies, How-to) 3. **Multi-language Output** - Generates content in both English and Vietnamese with customizable tone 4. **Video Rendering** - Automatically converts content to videos and infographics using Remotion ## Installation ### Prerequisites ```bash node >= 18.0.0 npm >= 9.0.0 ``` ### Setup ```bash # Clone the repository git clone https://github.com/pennydinh/marketing-pineline-share.git cd marketing-pineline-share # Install dependencies npm install # Copy environment template cp .env.example .env ``` ### Environment Configuration Create a `.env` file with the following variables: ```bash # AI Services ANTHROPIC_API_KEY=your_claude_api_key_here OPENAI_API_KEY=your_openai_api_key_here # Research APIs RAPIDAPI_KEY=your_rapidapi_key_here TWITTER_BEARER_TOKEN=your_twitter_bearer_token_here # Database (if applicable) DATABASE_URL=your_database_url_here # Remotion Configuration REMOTION_AWS_ACCESS_KEY_ID=your_aws_access_key_here REMOTION_AWS_SECRET_ACCESS_KEY=your_aws_secret_key_here ``` ## Key Architecture ### Project Structure ``` marketing-pipeline-share/ ├── src/ │ ├── research/ # Web scraping and data collection │ ├── ai/ # AI content generation (Claude/OpenAI) │ ├── video/ # Remotion video rendering │ ├── utils/ # Shared utilities │ └── app/ # Next.js application ├── remotion/ # Remotion video templates └── public/ # Static assets ``` ## Core Workflows ### 1. Research & Data Collection #### Scraping Recent Content ```typescript import { ResearchEngine } from './src/research/engine'; // Initialize research engine const researcher = new ResearchEngine({ sources: ['techcrunch', 'a16z', 'twitter', 'linkedin'], timeframe: '24h', keywords: ['AI', 'marketing automation'] }); // Fetch and analyze data const insights = await researcher.gather(); console.log(insights); // { // articles: [...], // trends: [...], // keyInsights: [...], // dataPoints: [...] // } ``` #### Custom Source Configuration ```typescript import { SourceConfig } from './src/research/types'; const customSource: SourceConfig = { name: 'custom-blog', url: 'https://example.com/blog', selectors: { title: '.post-title', content: '.post-content', date: '.post-date' }, rateLimit: 1000 // ms between requests }; const researcher = new ResearchEngine({ customSources: [customSource] }); ``` ### 2. AI Content Generation #### Using Claude for Content Creation ```typescript import { ClaudeContentGenerator } from './src/ai/claude'; const generator = new ClaudeContentGenerator({ apiKey: process.env.ANTHROPIC_API_KEY, model: 'claude-3-opus-20240229' }); // Generate content from research data const content = await generator.create({ research: insights, format: 'toplist', language: 'en', tone: 'professional', targetAudience: 'marketers', wordCount: 1500 }); console.log(content); // { // title: "Top 10 AI Marketing Trends in 2024", // body: "...", // metadata: { ... } // } ``` #### OpenAI Integration ```typescript import { OpenAIContentGenerator } from './src/ai/openai'; const openaiGen = new OpenAIContentGenerator({ apiKey: process.env.OPENAI_API_KEY, model: 'gpt-4-turbo-preview' }); // Generate multiple format variations const variations = await openaiGen.createVariations({ baseContent: content, formats: ['how-to', 'case-study', 'pov'], count: 3 }); ``` #### Multi-language Generation ```typescript import { MultiLanguageGenerator } from './src/ai/multilang'; const mlGenerator = new MultiLanguageGenerator({ claudeKey: process.env.ANTHROPIC_API_KEY, openaiKey: process.env.OPENAI_API_KEY }); // Generate parallel English and Vietnamese content const bilingualContent = await mlGenerator.generateParallel({ research: insights, languages: ['en', 'vi'], format: 'toplist', maintainTone: true }); console.log(bilingualContent.en.title); console.log(bilingualContent.vi.title); ``` ### 3. Video Generation with Remotion #### Basic Video Rendering ```typescript import { bundle } from '@remotion/bundler'; import { renderMedia, selectComposition } from '@remotion/renderer'; import { ContentToVideoConverter } from './src/video/converter'; // Convert content to video props const converter = new ContentToVideoConverter(); const videoProps = converter.transform(content); // Bundle Remotion composition const bundled = await bundle({ entryPoint: './remotion/index.ts', webpackOverride: (config) => config }); // Get composition const composition = await selectComposition({ serveUrl: bundled, id: 'ContentVideo', inputProps: videoProps }); // Render video await renderMedia({ composition, serveUrl: bundled, codec: 'h264', outputLocation: `out/${content.title}.mp4`, inputProps: videoProps }); ``` #### Custom Video Template ```typescript // remotion/ContentVideo.tsx import { AbsoluteFill, useCurrentFrame, useVideoConfig } from 'remotion'; export const ContentVideo: React.FC<{ title: string; points: string[]; branding: BrandConfig; }> = ({ title, points, branding }) => { const frame = useCurrentFrame(); const { fps } = useVideoConfig(); return (

{title}

{points.map((point, idx) => (

(idx + 1) * fps ? 1 : 0, transition: 'opacity 0.5s' }} > {point}

))}
); }; ``` #### Platform-Specific Export ```typescript import { PlatformOptimizer } from './src/video/optimizer'; const optimizer = new PlatformOptimizer(); // Render for multiple platforms const platforms = ['tiktok', 'reels', 'shorts']; for (const platform of platforms) { const config = optimizer.getConfig(platform); await renderMedia({ composition, serveUrl: bundled, codec: 'h264', outputLocation: `out/${platform}/${content.title}.mp4`, inputProps: videoProps, width: config.width, height: config.height, fps: config.fps }); } ``` ### 4. Complete Pipeline Automation ```typescript import { ContentPipeline } from './src/pipeline'; // Initialize full pipeline const pipeline = new ContentPipeline({ research: { sources: ['techcrunch', 'a16z'], timeframe: '24h' }, ai: { provider: 'claude', apiKey: process.env.ANTHROPIC_API_KEY, fallbackProvider: 'openai', fallbackKey: process.env.OPENAI_API_KEY }, video: { enabled: true, platforms: ['tiktok', 'reels', 'shorts'] }, output: { directory: './output', formats: ['markdown', 'html', 'json'] } }); // Execute full pipeline const results = await pipeline.execute({ keyword: 'AI marketing automation', contentFormats: ['toplist', 'how-to'], languages: ['en', 'vi'], generateVideos: true }); console.log(results); // { // research: { ... }, // content: [ // { format: 'toplist', language: 'en', ... }, // { format: 'toplist', language: 'vi', ... }, // { format: 'how-to', language: 'en', ... } // ], // videos: [ // { platform: 'tiktok', path: '...' }, // { platform: 'reels', path: '...' } // ] // } ``` ## API Reference ### ResearchEngine ```typescript class ResearchEngine { constructor(config: ResearchConfig); gather(): Promise; addSource(source: SourceConfig): void; setTimeframe(timeframe: string): void; } ``` ### ClaudeContentGenerator ```typescript class ClaudeContentGenerator { constructor(config: ClaudeConfig); create(params: ContentParams): Promise; refine(content: string, instructions: string): Promise; } ``` ### ContentPipeline ```typescript class ContentPipeline { constructor(config: PipelineConfig); execute(params: ExecutionParams): Promise; schedule(params: ScheduleParams): Promise; } ``` ## Common Patterns ### Pattern 1: Daily Content Automation ```typescript import cron from 'node-cron'; // Schedule daily content generation at 6 AM cron.schedule('0 6 * * *', async () => { const pipeline = new ContentPipeline(defaultConfig); const results = await pipeline.execute({ keyword: 'trending tech news', contentFormats: ['toplist'], languages: ['en'], generateVideos: true }); // Auto-publish to platforms await publishToWordPress(results.content[0]); await uploadToYouTube(results.videos[0]); }); ``` ### Pattern 2: Multi-Topic Content Generation ```typescript const topics = [ 'AI marketing trends', 'Social media automation', 'Content creation tools' ]; const batchResults = await Promise.all( topics.map(topic => pipeline.execute({ keyword: topic, contentFormats: ['how-to'], languages: ['en', 'vi'] }) ) ); ``` ### Pattern 3: Custom AI Prompt Engineering ```typescript import { PromptTemplate } from './src/ai/prompts'; const customTemplate = new PromptTemplate({ system: `You are an expert marketing content writer specializing in ${industry}.`, user: `Create a ${format} article about ${topic} targeting ${audience}. Include: - Data-backed insights from recent research - Actionable takeaways - SEO-optimized structure Research data: ${researchData}` }); const content = await generator.create({ prompt: customTemplate.render({ industry: 'SaaS', format: 'case-study', topic: insights.mainTrend, audience: 'startup founders', researchData: JSON.stringify(insights) }) }); ``` ## Troubleshooting ### Issue: API Rate Limits ```typescript // Implement exponential backoff import { retry } from './src/utils/retry'; const content = await retry( () => generator.create(params), { maxAttempts: 3, delayMs: 1000, exponentialBackoff: true } ); ``` ### Issue: Video Rendering Memory Issues ```typescript // Use concurrency limits import pLimit from 'p-limit'; const limit = pLimit(2); // Max 2 concurrent renders const videoPromises = platforms.map(platform => limit(() => renderForPlatform(platform, content)) ); await Promise.all(videoPromises); ``` ### Issue: Scraping Failures ```typescript // Add fallback sources const researcher = new ResearchEngine({ sources: ['techcrunch', 'a16z'], fallbackSources: ['medium', 'dev.to'], onSourceFail: (source, error) => { console.warn(`Source ${source} failed:`, error); }, minSourcesRequired: 1 }); ``` ### Issue: Content Quality Validation ```typescript import { ContentValidator } from './src/utils/validator'; const validator = new ContentValidator({ minWordCount: 1000, requireHeadings: true, checkReadability: true }); const content = await generator.create(params); if (!validator.validate(content)) { // Regenerate with refined prompt content = await generator.refine(content, validator.getSuggestions()); } ``` ## CLI Usage If the project includes CLI commands: ```bash # Generate content from keyword npm run generate -- --keyword "AI trends" --format toplist --lang en # Run full pipeline npm run pipeline -- --config ./config/production.json # Render video only npm run render -- --input ./content/article.json --platform reels # Schedule automated runs npm run schedule -- --cron "0 6 * * *" --topics ./topics.json ``` ## Best Practices 1. **Always use environment variables** for API keys 2. **Implement rate limiting** when scraping external sources 3. **Cache research data** to avoid redundant API calls 4. **Validate content quality** before rendering videos 5. **Use concurrency limits** for video rendering to manage resources 6. **Set up error monitoring** for production pipelines 7. **Version control your prompt templates** for reproducible results ## Resources - GitHub: https://github.com/pennydinh/marketing-pineline-share - Remotion Docs: https://remotion.dev - Anthropic Claude API: https://docs.anthropic.com - OpenAI API: https://platform.openai.com/docs

相关技能

媒体内容

Nano Banana Pro

以纳米·香蕉Pro(Gemini 3 Pro Image)来生成/编辑图像. 用于创建/修改请求,包括编辑。 支持文本到图像+图像到图像; 1K/2K/4K; 使用 -- input-image.

媒体内容

Youtube Watcher

从YouTube视频中获取并读取文字记录. 需要汇总视频时使用,回答有关视频内容的问题,或者从中提取信息.

媒体内容

Atxp

Access ATXP支付API工具用于网页搜索,AI图像生成,音乐创建,视频生成,X/Twitter搜索,电子邮件,代理账户管理. 当用户需要实时网络搜索时使用,AI生成的媒体(图像,音乐,视频),X/Twitter搜索,发送/接收电子邮件,或创建和资助代理账户. 需要通过“ …

媒体内容

Youtube

YouTube Data API与管理的OAuth的集成. 搜索视频,管理播放列表,访问频道数据,并与评论互动. 用户想与YouTube互动时使用此技能. 对于其他第三方应用,使用api-gateway技能(https://clawhub.ai/byungkyu/api-gate…