#36 · Primary category: Writing & Content

ai-trend-publish

ai weixin

TrendPublish: Full-auto AI content generation & publishing for WeChat, multi-source scraping (Twitter/X, websites), smart analysis/ranking, scheduled posts, multi-template, using DeepseekAI, Qianwen, Xunfei models; Node.js/TypeScript.

Project last updated:08/04/26

GitHub Stars

3.2K

Forks

439

Contributors

8

License

MIT

Why we included this project

Anyone who runs a daily-publishing WeChat official account knows the real bottleneck: turning a pile of feeds into one reliable, well-written article without burning hours. TrendPublish automates that pipeline by pulling candidates from sources like RSS, Hacker News, arXiv, and Twitter/X, then using LLMs to cluster and rank them, fill in missing evidence, draft the headline and body, review quality, and render WeChat-compatible HTML. What sets it apart from an RSS summarizer is the emphasis on reviewability: every run leaves step timelines, quality reports, rendered HTML, images, and config snapshots, so you can see why a piece was chosen and where to step in. A dry-run mode previews everything before anything reaches WeChat, the dashboard handles multi-account publishing with per-account tone and positioning, and it runs locally, in Docker, or on Cloudflare Workers.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category