This project provides an end-to-end AI-powered workflow for generating WeChat official account articles, automating tasks from trending topic discovery to draft submission. It targets users of Claude Code, content creators, and marketers seeking to streamline article production with AI assistance for writing, visuals, and SEO. The primary benefit is a significant reduction in manual effort for a complete article publishing pipeline.

How It Works

The project functions as a Claude Code Skill, orchestrating a series of Python scripts and LLM integrations. It begins by fetching real-time trending topics from platforms like Weibo, Toutiao, and Baidu. Subsequently, it analyzes keywords for SEO potential, generates multiple topic ideas and writing frameworks, and then drafts the article, focusing on removing AI traces and adapting to specific client styles. AI is also employed for generating cover and internal images using providers like Doubao or OpenAI. Finally, it converts Markdown content to WeChat-compatible inline HTML, publishes the draft to WeChat, and can even learn writing styles from manual edits for continuous improvement.

Quick Start & Requirements

Installation involves adding the repository as a Claude Code Skill path or copying it into the ~/.claude/skills/ directory. Python dependencies are installed via pip install -r requirements.txt after navigating to the wewrite directory. Configuration requires editing config.yaml to include WeChat official account appid/secret for publishing and statistics, along with API keys for image generation providers (Doubao-Seedream or OpenAI DALL-E).

Highlighted Details

• Comprehensive Workflow: Automates hot topic grabbing, SEO analysis, topic/framework generation, AI writing, AI image creation, Markdown-to-WeChat HTML conversion, and draft publishing.
• AI-Driven Content: Leverages LLMs for sophisticated content generation, style adaptation, and SEO optimization.
• Flexible Image Generation: Supports both Doubao (via Ark API) and OpenAI (DALL-E 3) for generating cover and internal article images.
• Customizable Formatting: Includes a theme system with multiple pre-built styles (e.g., professional-clean, tech-modern) for converting Markdown to WeChat-specific inline HTML.
• Style Learning: Features a learn_edits.py script to extract writing style patterns from manual revisions, enabling better client-specific content adaptation.

Maintenance & Community

Specific details regarding project maintainers, community channels (like Discord or Slack), or active sponsorships are not detailed in the provided documentation.

Licensing & Compatibility

The project is released under the MIT License. This license is permissive and generally allows for commercial use, modification, and distribution, including integration within closed-source applications.

Limitations & Caveats

Configuration requires sensitive API keys for WeChat publishing and AI image generation, which may incur usage costs. The effectiveness of AI-generated content and images is dependent on the quality of prompts and the underlying models. The primary intended usage is within the Claude Code environment, implying a dependency on that platform.