canghe-skills  by freestylefly

AI toolkit for advanced content generation and web utility

Created 3 months ago
289 stars

Top 90.9% on SourcePulse

GitHubView on GitHub
Project Summary

Summary

This repository, freestylefly/canghe-skills, provides a comprehensive suite of "skills" for Claude Code users, enhancing daily work efficiency. It offers tools for content creation, AI generation, and utility tasks, aiming to streamline workflows.

How It Works

Skills integrate into Claude Code as plugins, installable via npx or the marketplace. Each skill leverages specific APIs (OpenAI, Google, DashScope, Volcengine) or browser automation (Chrome CDP) for its functions. This modular approach allows users to add capabilities for content generation, AI media creation, and data processing.

Quick Start & Requirements

  • Prerequisites: Node.js environment, npx bun command capability.
  • Installation: Recommended: npx skills add freestylefly/canghe-skills. Alternatives include Claude Code's /plugin marketplace add or direct /plugin install commands.
  • Configuration: Many skills require API keys (OpenAI, Google, DashScope, Volcengine) or environment variables, configurable via .env files.

Highlighted Details

  • Content Creation: Tools for infographics (canghe-xhs-images, canghe-infographic), cover images (canghe-cover-image), slide decks (canghe-slide-deck), comics (canghe-comic), manga dramas/videos (canghe-manga-drama, canghe-manga-style-video), and article illustrations (canghe-article-illustrator).
  • AI Generation: Image generation (canghe-image-gen) via multiple providers, Gemini Web text/image generation (canghe-danger-gemini-web), AI video creation (canghe-seedance-video), and digital human avatar videos (flyworks-avatar-video).
  • Utility Tools: Markdown conversion from URLs (canghe-url-to-markdown) and X/Twitter (canghe-danger-x-to-markdown), WeChat article extraction (canghe-wechat-article-extractor), image compression (canghe-compress-image), markdown formatting (canghe-format-markdown), video analysis (canghe-volcengine-video-understanding), and document parsing (paddleocr-doc-parsing).
  • Social Media: Skills for posting to X (canghe-post-to-x) and WeChat Official Accounts (canghe-post-to-wechat).

Maintenance & Community

Skills are adapted from JimLiu/baoyu-skills. The repository recommends maintaining personal skills repositories. Specific community links or active maintainer details are not provided.

Licensing & Compatibility

Licensed under MIT, permitting commercial use and integration into closed-source projects.

Limitations & Caveats

canghe-danger-gemini-web and canghe-danger-x-to-markdown use reverse-engineered, unofficial APIs, posing risks of breakage or account restrictions. Some features require API keys. canghe-post-to-wechat offers API and browser methods, with the latter needing manual authentication.

Health Check
Last Commit

1 month ago

Responsiveness

Inactive

Pull Requests (30d)
0
Issues (30d)
0
Star History
75 stars in the last 30 days

Explore Similar Projects

Feedback? Help us improve.