45 lines
2.2 KiB
Markdown
45 lines
2.2 KiB
Markdown
---
|
||
title: "Marketing Carousel Growth Engine"
|
||
type: source
|
||
tags: [marketing, ai-agent, tiktok, instagram, carousel, the-agency]
|
||
sources: []
|
||
last_updated: 2026-04-21
|
||
---
|
||
|
||
## Source File
|
||
- [[raw/Agent/agency-agents/marketing/marketing-carousel-growth-engine.md]]
|
||
|
||
## Summary
|
||
- 核心主题：TikTok/Instagram 轮播内容自动生成与发布系统
|
||
- 问题域：社交媒体增长运营、内容自动化生产
|
||
- 方法/机制：Playwright 网站分析 → Gemini 图片生成 → Upload-Post API 发布 → 数据分析反馈循环
|
||
- 结论/价值：实现每日自主轮播内容生产，通过数据驱动迭代优化
|
||
|
||
## Key Claims
|
||
- 6-slide 叙事结构（Hook → Problem → Agitation → Solution → Feature → CTA）确保内容转化
|
||
- Gemini image-to-image 保证 6 张幻灯片视觉一致性
|
||
- Upload-Post API 同时发布到 TikTok 和 Instagram
|
||
- 学习系统持续积累最佳 hook、发布时间、视觉风格
|
||
|
||
## Key Quotes
|
||
> "You never ask for permission between steps — you research, generate, verify, publish, and learn, then report back with results." — 自主执行原则
|
||
|
||
## Key Concepts
|
||
- [[轮播叙事结构 (Carousel Narrative Arc)]]：6-slide 标准结构（Hook → Problem → Agitation → Solution → Feature → CTA）
|
||
- [[视觉一致性引擎 (Visual Coherence Engine)]]：Slide 1 建立视觉 DNA，Slides 2-6 通过 Gemini image-to-image 保持一致性
|
||
- [[分析反馈循环 (Analytics Feedback Loop)]]：Fetch → Extract → Update → Plan 持续迭代
|
||
- [[自主质量保证 (Autonomous QA)]]：视觉模型验证每张幻灯片，失败仅重生成问题幻灯片
|
||
|
||
## Key Entities
|
||
- [[Gemini API]]：Google 图片生成服务，`gemini-3.1-flash-image-preview` 模型
|
||
- [[Upload-Post API]]：社交媒体发布与数据分析 API
|
||
- [[Playwright]]：网站抓取与内容分析工具
|
||
- [[The Agency]]：开源 AI Agent 集合项目
|
||
|
||
## Connections
|
||
- [[Marketing Carousel Growth Engine]] ← uses ← [[Playwright]]（网站分析）
|
||
- [[Marketing Carousel Growth Engine]] ← uses ← [[Gemini API]]（图片生成）
|
||
- [[Marketing Carousel Growth Engine]] ← uses ← [[Upload-Post API]]（发布与分析）
|
||
- [[Marketing Carousel Growth Engine]] → feeds_into → [[Analytics Feedback Loop]]（数据驱动优化）
|
||
|
||
## Contradictions |