miantiao-me/aigc-weekly

Agili 的 AIGC 周刊 - 一个由 Agentic AI Agent 驱动的 AIGC（人工智能生成内容）精选周刊。

What it solves

Keeping up with the fast-moving world of AI-generated content (AIGC) is time-consuming. This project automates the curation of a weekly newsletter that collects the latest news, tools, and resources, so readers can get a digest without manually scouring the web. It also shows how AI agents can work with a content management system to publish posts automatically.

How it works

The project is built around three main parts:

  1. A web frontend and CMS – A Next.js app (using App Router) serves the public reader pages and provides an admin interface via Payload CMS. The CMS stores the newsletter posts and manages content structure.
  2. An AI agent service – An OpenCode agent runs on Cloudflare Containers. It uses tools like Firecrawl to scrape and extract information from websites, then uses the Model Context Protocol (MCP) to talk directly to the CMS and update it with curated items. This is what makes the curation autonomous.
  3. A Cloudflare Worker – This acts as a gateway that forwards requests to the agent container and controls the container's lifecycle, allowing the agent to start on demand and stay serverless.

All data is stored in Cloudflare D1 (SQLite database), and files are stored in Cloudflare R2 (object storage). The whole stack runs on Cloudflare's edge infrastructure, which gives high performance and low latency.

Who it’s for

  • Readers who want a regular, AI-curated digest of AIGC news, tools, and resources via the website or RSS feed.
  • Developers interested in combining AI agents, serverless architecture, and headless CMS systems. The project provides a reference implementation for building similar automated content pipelines.

Highlights

  • Automated curation – An AI agent discovers and filters content without manual editorial effort.
  • Modern tech stack – Built with Next.js 15, Payload CMS 3.0, OpenNext, and Cloudflare edge services.
  • Fully serverless – Runs entirely on Cloudflare Workers, D1, R2, and Containers, providing scalability and low operating overhead.
  • Agent-MCP integration – Uses the Model Context Protocol so the AI agent can interact directly with the CMS, enabling seamless content updates.
  • RSS feed – Provides an RSS subscription channel for readers who prefer feed readers.

Related

  • Project
    langgenius/dify-plugins

    A plugin ecosystem and marketplace for Dify, enabling developers to extend AI applications with custom models, tools, and agent reasoning strategies.

  • Project
    freddy-schuetz/n8n-claw

    A self-hosted AI agent built on n8n and PostgreSQL that integrates long-term memory, task management, and web automation via Telegram or HTTP API.

  • Project
    vakra-dev/reader

    Reader is an open‑source Node library/CLI that wraps Playwright to provide a production‑grade, stealthy web‑scraping engine. It handles browser pooling, anti‑bot bypass, proxy rotation, and returns clean markdown or HTML, making it useful for AI agents that need reliable web retrieval.

  • Project
    zgsfsec/Full-Chain-Sentinel-System

    Full Chain Sentinel System is an authorized penetration-testing agent that combines an AI assistant with per-project Kali containers, a bilingual web console, and tool integrations for asset discovery, web research, and mobile testing—built for security teams needing a controlled, audited, and multi-platform offensive security workflow.